System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving and analyzing a curriculum vitae; means for identifying sections of the curriculum vitae using natural-language processing techniques; means for generating a set of questions based on the analysis; means for conducting interviews using video conferencing tools and reading out the questions; means for converting audio responses to text and analyzing the text using natural-language processing techniques; means for evaluating the analysis and generating feedback materials; and means for transmitting the feedback materials to interviewers.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Corporate recruitment activities require reading a large number of application forms and conducting multiple interviews, which incurs high costs. Furthermore, mismatches in recruitment can lead to early turnover and further waste of costs. Therefore, there is a need for an efficient and accurate method of conducting early-stage interviews. With conventional methods, the subjectivity and emotions of the interviewer can affect the evaluation, which is one of the causes of mismatches. Therefore, an innovative recruitment system is needed that can simultaneously reduce recruitment costs and prevent mismatches. [Means for solving the problem]

[0005] This invention solves the above-mentioned problems with a recruitment system that includes a means for receiving and analyzing resumes, a means for identifying each section of the resume using natural language processing technology, a means for generating a question set based on the analysis results, a means for conducting interviews using a video conferencing tool and reading out the questions, a means for converting voice responses into text and analyzing them using natural language processing technology, a means for evaluating the analysis results, creating feedback materials, and a means for sending the feedback materials to interviewers. This system enables AI to efficiently and objectively conduct interviews at the early stage of recruitment, reducing recruitment costs and preventing mismatches. Furthermore, by including a means for conducting knowledge tests in a predetermined format and evaluating responses, and a means for notifying candidates of the interview schedule and method, a more accurate recruitment process can be provided.

[0006] A "resume" is a document in which an applicant lists information such as their past educational background, work history, skills, and motivation for applying.

[0007] "Natural language processing technology" is a technology that enables computers to analyze, understand, and generate human language.

[0008] A "question set" is a collection of questions to be asked of an applicant during an interview, and is generated based on the analyzed resume information.

[0009] A "video conferencing tool" is software and a platform that allows multiple participants to communicate via video and audio in real time over the Internet.

[0010] "Speech generation AI" is a technology or system that uses artificial intelligence to generate human-like speech from text data.

[0011] "Speech recognition technology" is a technology that converts voice data into text data.

[0012] A "feedback document" is a report that summarizes the results and evaluation of an interview and is provided to the next interviewer, including the applicant's strengths and weaknesses and any additional points that need to be checked.

[0013] A "knowledge test" is a test to evaluate an applicant's specialized knowledge and skills.

[0014] An "assessment tool" is a method or system for analyzing an applicant's interview or knowledge test responses and scoring them based on certain criteria.

[0015] "Notification method" refers to the method used to inform applicants of the interview date and method, such as email or messaging apps. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] System Overview

[0038] This invention is a system that uses artificial intelligence (generative AI models) to conduct interviews at the early stage of the recruitment process, reducing recruitment costs and preventing mismatches.

[0039] Resume analysis

[0040] The server receives the resume uploaded by the candidate and converts it into text data using optical character recognition (OCR) technology, then uses natural language processing technology to identify sections of the resume and extract information such as educational background, work history, skills, and motivation for applying.

[0041] Question Set Generation

[0042] Based on the extracted information, the server generates a set of interview questions for the candidate, which are customized according to the company's requirements and the candidate's characteristics.

[0043] Conducting interviews

[0044] The device notifies the candidate of the date, time, and method of the interview. The candidate will be interviewed at the specified date and time using a video conferencing tool (e.g., Zoom). Questions are read aloud by a voice generation AI, and the user (candidate) answers them.

[0045] Analysis of responses

[0046] The server converts the candidate's voice responses into text in real time, which is then analyzed using natural language processing technology to evaluate the specificity and logic of the responses.

[0047] Knowledge test

[0048] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal, and the candidate's answers are sent to the server for evaluation.

[0049] Creating and Providing Feedback

[0050] Based on the results of the interview and knowledge test, the server creates feedback materials, including the candidate's strengths and weaknesses and points to check next time. The created feedback materials are sent to the first interviewer.

[0051] Specific examples of processing

[0052] For example, when a software engineer applicant uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as, "Please tell us about your specific project experience using Python and Django" are generated. On the day of the interview, the voice generation AI on the device reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0053] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0054] The processing flow will be explained below.

[0055] Step 1:

[0056] The server receives resume files uploaded by applicants, which are assumed to be in PDF or Word format.

[0057] Step 2:

[0058] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[0059] Step 3:

[0060] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Specifically, it uses a text analysis algorithm.

[0061] Step 4:

[0062] The server generates a set of questions based on the extracted information, taking into account questions relevant to the position, such as "Tell me about your Python programming experience."

[0063] Step 5:

[0064] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link). Notifications will be sent via email or messaging system.

[0065] Step 6:

[0066] Users are interviewed at the appointed time using a video conferencing tool (such as Zoom), and the device reads out pre-generated questions using speech generation AI.

[0067] Step 7:

[0068] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[0069] Step 8:

[0070] The server converts the user's response voice into text data using speech recognition technology, and the converted text data is analyzed using natural language processing technology.

[0071] Step 9:

[0072] The server evaluates the analyzed answer text and assigns a score based on points such as specificity and logic.

[0073] Step 10:

[0074] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[0075] Step 11:

[0076] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[0077] Step 12:

[0078] The server evaluates the candidate's knowledge test answers and calculates a score.

[0079] Step 13:

[0080] The server combines the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what needs to be checked next time.

[0081] Step 14:

[0082] The server sends the created feedback material to the first interviewer, who uses the material to prepare for the next interview.

[0083] These are the specific steps for efficiently conducting the recruitment process. This system reduces recruitment costs and enables the selection of appropriate personnel.

[0084] Example 1

[0085] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0086] A problem with the traditional recruitment process is that the initial interview stage takes a lot of time and costs a lot of money. Furthermore, evaluations often depend on the interviewer's subjective judgment, making it difficult to objectively evaluate the candidate's aptitude and abilities. For these reasons, there is a need to prevent mismatches and streamline the recruitment process.

[0087] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0088] In this invention, the server includes means for receiving and analyzing resumes, means for converting resumes into text data using optical character recognition technology, means for identifying each section of the resume and extracting information using natural language processing technology, means for generating a set of interview questions based on the extracted information, means for customizing the questions according to the company's requirements and the applicant's characteristics, means for notifying the applicant of the interview date and method, means for conducting the interview using a video conferencing tool and reading the questions aloud using speech generation technology, means for converting the user's voice responses into text, means for analyzing and evaluating the converted text using natural language processing technology, means for creating feedback materials based on the results of the interview and knowledge test, and means for sending the feedback materials to the interviewer, thereby enabling an efficient hiring process and objective evaluation.

[0089] "Means for receiving and analyzing resumes" refers to a series of functions for receiving resumes from applicants and analyzing their contents.

[0090] "Optical character recognition technology" is a technology that extracts character information from documents such as images and PDFs.

[0091] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[0092] "Means for identifying each section of a resume and extracting information" refers to a function that analyzes the contents of a resume to identify each section, such as educational background, work history, and skills, and extracts that information.

[0093] The "means for generating a set of interview questions" is a function that creates questions for candidates based on the analyzed resume information.

[0094] "A means of customizing questions according to the company's requirements and the characteristics of the applicant" refers to a function that appropriately changes questions based on the skills the company is looking for and the background of the applicant.

[0095] "Means of notifying applicants of the interview date and method" is a function that informs applicants of the date and time of the interview and how it will be conducted.

[0096] "Means for conducting interviews using video conferencing tools and reading questions aloud using voice generation technology" refers to a function that conducts interviews using video conferencing software and presents generated questions to candidates as audio.

[0097] The "means for converting the user's voice response into text" is a function for converting the content of the user's verbal response into text data in real time.

[0098] "Means for analyzing and evaluating the converted text using natural language processing technology" refers to a function that analyzes text converted from speech and evaluates the specificity and logic of the content.

[0099] The "means for creating feedback materials" is a function for creating feedback documents for applicants based on the result data of the interview and knowledge test.

[0100] The "means of sending feedback materials to interviewers" is a function for sending the created feedback materials to the first interviewer and other related staff.

[0101] This invention relates to a system that uses artificial intelligence (generative AI models) to conduct interviews in the early stages of the recruitment process, which reduces recruitment costs and prevents mismatches.

[0102] System Overview

[0103] The system is designed to automate the analysis of resumes, generation of question sets, conducting interviews, analyzing responses, conducting knowledge tests, and creating and providing feedback. Each function is described in detail below.

[0104] Resume analysis

[0105] 1. A user accesses the system and uploads a resume.

[0106] 2. The server receives the resume and converts it into text data using optical character recognition (OCR) technology. Specifically, it uses Tesseract OCR.

[0107] 3. The server analyzes the converted text data using natural language processing technology (e.g., spaCy), identifies each section of the resume (educational history, work history, skills, motivation, etc.), and extracts the necessary information.

[0108] Question Set Generation

[0109] 1. The server generates a set of interview questions based on the extracted information, customizing the questions according to the company's requirements and the applicant's characteristics.

[0110] 2. For example, the server uses a generative AI model such as GPT-3 to create questions such as, "Tell me about your specific project experience using Python and Django."

[0111] Conducting interviews

[0112] 1. The device notifies candidates of the date, time, and method of their interview via email, SMS, or a dedicated app.

[0113] 2. Candidates will be interviewed via video conferencing tools (e.g., Zoom) at the appointed date and time.

[0114] 3. During the interview, a voice generation AI (e.g., Google Text-to-Speech) on the device reads out the questions retrieved from the server.

[0115] Analysis of responses

[0116] 1. The candidate answers and the device records the audio.

[0117] 2. The server converts the recorded audio into text in real time using Google Speech-to-Text.

[0118] 3. The converted text is analyzed using natural language processing technology (e.g., spaCy) to evaluate the specificity and logic of the answers.

[0119] Knowledge test

[0120] 1. The server generates knowledge test questions and presents them to the candidate in a predetermined format. The test content includes coding questions and theory questions.

[0121] 2. The candidate answers, and the answer is sent to the server for evaluation.

[0122] Creating and Providing Feedback

[0123] 1. The server creates feedback materials based on the results of the interview and knowledge test. The feedback materials include the candidate's strengths and weaknesses and what needs to be checked next time.

[0124] 2. The prepared feedback materials will be sent to the first interviewer.

[0125] Specific examples

[0126] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as "Tell us about your specific project experience using Python and Django" are generated. On the day of the interview, a voice generation AI reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0127] Example prompt sentence:

[0128] "Design a system that parses resumes uploaded by candidates, generates interview questions, conducts interviews, parses responses, administers knowledge tests, and generates feedback materials."

[0129] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0130] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0131] Step 1:

[0132] A user accesses the system and uploads a resume. The input is the user's resume file, and the output is the resume data stored on the server. Specifically, the user creates a profile through the web interface and clicks the resume upload button. The server receives this upload request and saves the resume.

[0133] Step 2:

[0134] The server receives the resume and converts it into text data using optical character recognition (OCR) technology. The input is the uploaded resume file, and the output is the converted text data. Specifically, the server invokes Tesseract OCR to extract text from the resume image.

[0135] Step 3:

[0136] The server uses natural language processing technology to analyze resumes and extract the necessary information. The input is text data, and the output is extracted information such as educational background, work history, skills, and motivation for applying. The server uses spaCy and NLTK to analyze the text and identify keywords and sections.

[0137] Step 4:

[0138] The server generates a set of interview questions based on the extracted information. The input is the extracted information, and the output is the generated set of questions. Specifically, the server retrieves requirements from the company's database and creates questions using a generative AI model (e.g., GPT-3).

[0139] Step 5:

[0140] The server customizes the questions according to the company's requirements and the applicant's characteristics. The input is the generated question set and the company's requirements, and the output is the customized question set. Specifically, the server performs filtering and text fine-tuning on the generated questions.

[0141] Step 6:

[0142] The device notifies the candidate of the interview date, time, and method. The input is information about the interview date, time, and method, and the output is the notified candidate. The device sends this information to the candidate via email, SMS, or a dedicated app.

[0143] Step 7:

[0144] A candidate will be interviewed at a specified date and time using a video conferencing tool. The input is the interview link and date and time information, and the output is joining the video conference. Specifically, the user will log in to the video conferencing tool (e.g., Zoom) using their device.

[0145] Step 8:

[0146] The device uses speech generation AI to read out questions. The input is a customized set of questions, and the output is audio questions. Specifically, the device uses the Google Text-to-Speech API to convert the questions into audio and present them to the candidate as audio played through the speaker.

[0147] Step 9:

[0148] The user (candidate) answers questions, and the device records the audio. The input is the user's voice response, and the output is the recorded audio data. Specifically, the device captures the audio using Zoom's recording function or a dedicated app.

[0149] Step 10:

[0150] The server converts the user's voice response into text. The input is the recorded voice data, and the output is the converted text data. Specifically, the server calls Google Speech-to-Text to convert the voice to text.

[0151] Step 11:

[0152] The server analyzes the converted text and evaluates the specificity and logic of the answer. The input is text data, and the output is the analysis result. Specifically, the server uses spaCy to analyze the text and evaluates it based on specific criteria.

[0153] Step 12:

[0154] The server generates knowledge test questions and presents them to candidates through their terminals. The input is the knowledge test requirements, and the output is a set of knowledge test questions. Specifically, the server uses a question generation algorithm to create coding questions and theory questions and display them on the terminal.

[0155] Step 13:

[0156] The candidate's answers are sent through the terminal and evaluated by the server. The input is the candidate's answer data, and the output is the evaluation result. Specifically, the terminal sends the answer data to the server, which then automatically scores it.

[0157] Step 14:

[0158] The server creates feedback documents based on the results of the interview and knowledge test. The input is the result data of the interview and knowledge test, and the output is the feedback documents. Specifically, the server aggregates the results and generates the feedback documents using a generative AI model.

[0159] Step 15:

[0160] The created feedback material is sent to the first interviewer. The input is the feedback material, and the output is the feedback sent to the first interviewer. In concrete terms, the server sends the created feedback material to the first interviewer via email or an internal system.

[0161] (Application example 1)

[0162] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0163] Maintenance work for factory robots is often complex and requires specialized knowledge, making the judgment of maintenance workers essential. However, because maintenance depends on the skills and experience of the workers, there is a problem of inconsistency in the quality of maintenance. Another issue is the lack of a means to efficiently provide work procedures and provide appropriate feedback in real time, which reduces work efficiency.

[0164] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0165] In this invention, the server includes means for receiving and analyzing a resume, means for identifying each section of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing tool and reading out the questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for sending the feedback materials to the interviewer, means for extracting text information from an image using optical character recognition technology, means for generating a maintenance procedure tailored to a specific situation using a generative artificial intelligence model, and means for evaluating the maintenance work based on the procedure, thereby enabling the efficiency and quality of robot maintenance work in a factory to be improved.

[0166] "Means for receiving and analyzing resumes" refers to a function for electronically receiving resumes submitted by applicants and analyzing their contents.

[0167] "Means for identifying each section of a resume using natural language processing technology" is a function that uses natural language processing technology to classify and identify the contents of a resume into sections such as educational history, work history, and skills.

[0168] The "means for generating a question set based on the analysis results" is a function for generating appropriate interview questions using the analysis results of a resume.

[0169] The "means for conducting an interview using a video conferencing tool and reading out questions" is a function for conducting an interview using a video conferencing tool and reading out the generated questions aloud.

[0170] "Means for converting voice responses into text and analyzing it using natural language processing technology" refers to a function for converting the applicant's voice responses into text data and analyzing that text using natural language processing technology.

[0171] The "means for evaluating the analysis results and creating feedback materials" is a function for evaluating the analysis results of the voice responses and creating materials summarizing feedback to the applicant.

[0172] The "means for sending feedback materials to interviewers" is a function for electronically sending the created feedback materials to interviewers.

[0173] "Means for extracting character information from an image using optical character recognition technology" is a function for reading character information in an image using optical character recognition technology and converting it into text data.

[0174] "Means for generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model" is a function for automatically generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model.

[0175] The "means for evaluating maintenance work based on the procedure" is a function for evaluating work performed based on the generated maintenance procedure.

[0176] This invention relates to a system for improving the efficiency and quality of robot maintenance work in a factory. In one embodiment of the present invention, how a server, a terminal, and a user work together to realize the system will be described.

[0177] Server processing

[0178] Resume analysis

[0179] The server receives the resume uploaded by the user (worker) and converts it into text data using optical character recognition (OCR) technology. The specific software used is OpenCV and pytesseract. After this, each section of the resume (educational history, work history, skills, etc.) is identified using natural language processing technology. The specific software used in this step is Python and a library for natural language processing.

[0180] Question Set Generation

[0181] The server generates a set of interview questions based on the analysis results. This part uses a generative AI model (e.g., GPT-3.5) to create questions that match the characteristics of the candidate and the company's requirements. These questions will also be used to generate maintenance procedures in the future.

[0182] Processing by the terminal

[0183] Conducting interviews

[0184] The device notifies the user of the date, time, and method of the interview. The user then attends the interview at the specified time using a video conferencing tool (e.g., Zoom). The questions are read aloud by a speech generation AI. This part uses Google Cloud Speech-to-Text and Text-to-Speech services for speech recognition and generation technology.

[0185] Analysis of responses

[0186] The server converts the user's voice response into text in real time using Google Cloud Speech-to-Text, and the converted text is analyzed using natural language processing technology to evaluate the specificity and logic of the response.

[0187] Robot Maintenance

[0188] Extracting character information using optical character recognition technology

[0189] The server extracts the robot's state information from the images using optical character recognition techniques, specifically OpenCV and pytesseract.

[0190] Generate maintenance procedures

[0191] The server uses a generative AI model to generate maintenance procedures for specific situations. GPT-3.5 is used to generate these procedures. For example, if the robot's status image displays "Error message 72: Motor overheating," the following prompt sentence is input to the generative AI model:

[0192] The robot status is displayed as "Error message 72: Motor overheated". Please suggest the following maintenance steps:

[0193] Maintenance work evaluation

[0194] The server evaluates the work performed based on the generated maintenance procedure and creates feedback materials based on the evaluation results, which are provided to the workers and used to improve future maintenance work.

[0195] As a result, the efficiency and quality of robot maintenance work within the factory will be improved.

[0196] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0197] Step 1:

[0198] The server receives resumes uploaded by users. The input is the resume uploaded by the user in PDF or image format. The server converts this resume into text data using optical character recognition (OCR) technology. Specifically, it uses OpenCV and pytesseract to recognize characters in the image and extract them as text data. The output is the text data of the resume.

[0199] Step 2:

[0200] The server analyzes the text data extracted by OCR using natural language processing technology. The input is the text data of a resume. The server analyzes this data and identifies each section, such as educational history, work history, and skills. Specifically, it uses Python and natural language processing libraries. The output is the text data of a resume with each section identified.

[0201] Step 3:

[0202] The server generates a set of interview questions based on the analysis results. The input is the text data for each identified section. The server uses a generative AI model (e.g., GPT-3.5) to create a question set that corresponds to the candidate's characteristics and the company's requirements. The output is the generated question set.

[0203] Step 4:

[0204] The terminal notifies the user of the date, time, and method of the interview. The input is the question set and interview schedule data generated from the server. The terminal sends the notification to the user using email or a messaging service. The output is data about the date, time, and method of the notification.

[0205] Step 5:

[0206] The user is interviewed at a specified date and time using a video conferencing tool (e.g., Zoom). The input is a set of questions generated by the server and the user's answers. A voice generation AI on the device reads the questions aloud. The output is the data of the user's voice responses.

[0207] Step 6:

[0208] The server converts the user's voice response into text data in real time. The input is the user's voice response. The server converts the voice to text data using Google Cloud Speech-to-Text technology. The output is the text data response.

[0209] Step 7:

[0210] The server analyzes the text data responses using natural language processing technology. The input is the text data responses. The server analyzes the converted text and evaluates the specificity and logic of the responses. The output is evaluation data of the analysis results.

[0211] Step 8:

[0212] The server creates feedback materials based on the analysis results. The input is the evaluation data. The server generates feedback sentences using a Python script. The output is the feedback materials.

[0213] Step 9:

[0214] The server sends the created feedback materials to the interviewer. The input is the feedback materials. The server sends the materials via email. The output is a notification that the materials have been sent.

[0215] Step 10:

[0216] The server uses optical character recognition technology to extract robot status information from images. The input is image data showing the robot's status. The server uses OpenCV and pytesseract to convert the character information in the image into text data. The output is text data of the robot's status.

[0217] Step 11:

[0218] The server uses a generative artificial intelligence model to generate maintenance procedures tailored to specific situations. The input is text data of the robot's status. The server uses a generative AI model (e.g., GPT-3.5) to generate appropriate maintenance procedures. For example, the prompt sentence "The robot's status is displayed as 'Error message 72: Motor overheating'. Please suggest the next maintenance procedure." is input to the model. The output is the generated maintenance procedure.

[0219] Step 12:

[0220] The server evaluates the maintenance work based on the procedure. The input is the generated maintenance procedure and the maintenance work result data. The server evaluates the work performed according to the procedure and creates the results as feedback material. The output is the evaluation result of the maintenance work and the feedback material.

[0221] Step 13:

[0222] The server sends the created feedback material to the worker. The input is the feedback material. The server sends the material via email. The output is a notification that the material has been sent.

[0223] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0224] System Overview

[0225] This invention is a system that uses artificial intelligence (generative AI models) in the early stages of the hiring process. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate hiring process.

[0226] Resume analysis

[0227] The server receives resumes uploaded by candidates, converts the resume files (PDF or Word files) into text data using optical character recognition (OCR), and then uses natural language processing (NLP) to extract information such as educational background, work history, skills, and motivation for applying from the resume text.

[0228] Question Set Generation

[0229] The server generates a set of questions based on the extracted information, customized to include content relevant to the position being applied for. For example, a software engineer candidate might be asked questions like, "Tell me about your Python programming experience."

[0230] Conducting interviews

[0231] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). The candidate takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions. The user (candidate) answers them.

[0232] Incorporating an emotion engine

[0233] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is then sent to the server.

[0234] Analysis of response and sentiment data

[0235] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. The response is evaluated based on factors such as specificity and logic. Emotion data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed.

[0236] Knowledge test

[0237] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[0238] Creating and Providing Feedback

[0239] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[0240] Specific examples of processing

[0241] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. During the interview, they are asked questions such as, "Tell us about your specific project experience using Python and Django," and the emotion engine analyzes the user's facial expressions and tone of voice while answering. The server analyzes the voice responses and emotion data and makes an evaluation such as, "The answers were highly specific and emotionally confident." This evaluation is reflected in the feedback materials and sent to the first interviewer.

[0242] By incorporating emotion recognition technology, this system enables objective evaluation that is not influenced by the interviewer's subjectivity, contributing to the selection of more appropriate personnel. It also improves convenience for applicants by providing flexible interview opportunities.

[0243] The processing flow will be explained below.

[0244] Step 1:

[0245] The server receives resume files uploaded by applicants, which are expected to be in PDF or Word format.

[0246] Step 2:

[0247] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[0248] Step 3:

[0249] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data.

[0250] Step 4:

[0251] The server generates a set of questions based on the extracted information, including questions such as "Tell me about your Python programming experience."

[0252] Step 5:

[0253] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link) via email or messaging system.

[0254] Step 6:

[0255] The user will be interviewed at the appointed date and time using a video conferencing tool (Zoom), with the device reading out pre-generated questions using a speech-generating AI.

[0256] Step 7:

[0257] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[0258] Step 8:

[0259] During the interview, the device activates an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time, which is then sent to the server.

[0260] Step 9:

[0261] The server uses speech recognition technology to convert the user's voice response into text data. The converted text is then analyzed using natural language processing technology to evaluate points such as the specificity and logic of the response. At the same time, emotional data obtained from an emotion recognition engine is also analyzed.

[0262] Step 10:

[0263] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[0264] Step 11:

[0265] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[0266] Step 12:

[0267] The server evaluates the candidate's knowledge test answers and calculates a score.

[0268] Step 13:

[0269] The server aggregates all responses and emotional data to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. The emotional data indicates the candidate's emotional changes during the interview and their overall emotional state.

[0270] Step 14:

[0271] The server transmits the created feedback material to the first interviewer, who can use this material to efficiently prepare for the next interview.

[0272] Example 2

[0273] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0274] In the traditional hiring process, it is easily influenced by the interviewer's subjectivity, making it difficult to make an objective evaluation. Furthermore, there is no way to evaluate the candidate's emotional changes and facial expressions in real time, making it difficult to make an accurate and fair evaluation. Furthermore, the typical interview process is time-consuming, labor-intensive, and inefficient.

[0275] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0276] In this invention, the server includes means for receiving and analyzing resumes, means for converting resume files into text data using optical character recognition technology, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for inputting prompts and generating appropriate questions using a generative AI model, means for conducting interviews using a video conferencing tool and reading out questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for analyzing facial expressions and tone of voice during the interview using an emotion recognition engine and extracting emotion data, means for integrating the analysis results and the emotion data and evaluating them, means for creating feedback materials, and means for sending the feedback materials to interviewers. This enables objective and accurate evaluation, including changes in the emotions of candidates, and improves the efficiency and fairness of the hiring process.

[0277] "Means for receiving and analyzing resumes" refers to the device or software used to receive resume files uploaded by applicants and analyze their contents.

[0278] "Optical character recognition technology" is a technology that extracts text information from image data and is used to convert digital resume files into text data.

[0279] "Natural language processing technology" is a technology that uses computers to analyze and understand text written in natural language, and is used to extract necessary information from text data such as resumes.

[0280] A "means for generating a question set" is a device or software for generating appropriate interview questions based on the analyzed resume data.

[0281] A "generative AI model" is a trained artificial intelligence model that has the ability to generate natural language when given a prompt sentence.

[0282] A "prompt" is an instruction given to an AI model to perform a specific task.

[0283] A "video conferencing tool" is software for sharing audio and video in real time over the Internet.

[0284] A "means for converting voice responses to text" is a device or software that uses voice recognition technology to convert voice responses during an interview into text form.

[0285] An "emotion recognition engine" is a device or software that analyzes a person's emotional state from audio and video data and extracts emotional data in real time.

[0286] The "means for creating feedback materials" refers to a device or software for generating feedback materials summarizing the evaluation of candidates based on the results of the interview and knowledge test.

[0287] The "means for sending feedback materials to interviewers" refers to a device or software for automatically sending the created feedback materials to interviewers.

[0288] This invention is a system that combines artificial intelligence (generative AI models) and an emotion recognition engine to perform more accurate evaluations in the early stages of the hiring process. This system automates the processes of resume analysis, question generation, interviews, response analysis, knowledge testing, and feedback creation.

[0289] Hardware and software used

[0290] The system uses the following hardware and software:

[0291] 1. Server

[0292] A server with the storage and processing power to receive resumes

[0293] Optical Character Recognition (OCR) engine (e.g., Tesseract)

[0294] Natural Language Processing (NLP) engines (e.g., SpaCy, BERT)

[0295] Generative AI models (e.g., GPT-3)

[0296] Voice recognition technology (e.g., Google Speech-to-Text)

[0297] Report generation tools (e.g. Tableau)

[0298] 2. Terminal

[0299] Devices for real-time communication with users (PCs, smartphones, tablets, etc.)

[0300] Video conferencing tools (e.g., Zoom, Microsoft Teams)

[0301] Voice generation AI (e.g. Amazon Polly)

[0302] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[0303] 3. Users

[0304] Applicants and interviewers

[0305] Specific processing of the program

[0306] Resume upload and analysis

[0307] Users upload their resumes to the server, which converts the received resume files (PDF or Word files) into text data using an optical character recognition (OCR) engine. It then uses natural language processing (NLP) to extract information such as educational background, work history, and skills from the resume text data.

[0308] Question Set Generation

[0309] The server generates a set of relevant questions for the candidate based on the analyzed resume data. It uses a generative AI model (GPT-3) to generate appropriate questions. For example, the server inputs a prompt such as, "Generate interview questions based on the candidate's programming skills."

[0310] Preparing for and conducting interviews

[0311] The device notifies the candidate of the date, time, and method of the interview, sends a link to a video conferencing tool (such as Zoom), and the interview takes place at the specified date and time. During the interview, a speech-generation AI (Amazon Polly) is used to read out pre-generated questions.

[0312] Analysis by emotion engine

[0313] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice, extracting emotional data in real time and sending it to the server.

[0314] Analysis of response and sentiment data

[0315] The server converts the applicant's voice responses into text data using speech recognition technology and analyzes them using natural language processing technology. The server evaluates the specificity and logic of the responses. It also analyzes emotional data obtained from an emotion recognition engine to evaluate emotional changes.

[0316] Knowledge test

[0317] The server creates knowledge test questions using a generative AI model and presents them to the candidate via their device. For example, the candidate might enter a prompt such as, "Please create a question about data structures and algorithms." The candidate's answers are then sent to the server for evaluation.

[0318] Creating and Providing Feedback

[0319] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. This feedback includes emotional data and is evaluated in detail. The created feedback materials are sent to the first interviewer.

[0320] Examples of prompt statements

[0321] 1. "Please parse this resume and extract Python and Django project experience."

[0322] 2. "Generate questions to assess knowledge of data structures and algorithms."

[0323] 3. "Evaluate the specificity of interview responses and create feedback materials that analyze emotional data."

[0324] By integrating emotion recognition and AI technology, the system eliminates subjective interviewer evaluations, significantly improving the accuracy and efficiency of the hiring process, while providing candidates with a flexible interview experience for increased convenience.

[0325] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0326] Step 1:

[0327] Resume upload and analysis

[0328] The user uploads their resume to the system. The server stores the received resume file (PDF or Word file) in storage. Optical character recognition (OCR) technology is used to convert the file into text data. The input is a PDF or Word file, and the output is text data. The server launches an OCR engine (e.g., Tesseract) to analyze the file and extract the text. Next, natural language processing (NLP) technology is used to extract information such as education, work history, and skills from the resume text data. An NLP engine (e.g., SpaCy) is used to identify the information for each section.

[0329] Step 2:

[0330] Question Set Generation

[0331] The server generates a set of questions relevant to the candidate based on the analyzed information. It uses a generative AI model (e.g., GPT-3) and inputs a prompt to generate questions. The input is the analyzed resume data and the prompt, and the output is a set of interview questions. For example, the prompt "Generate interview questions based on the candidate's programming skills" is input to the generative AI model. The generated questions are saved for interview preparation.

[0332] Step 3:

[0333] Preparing for and conducting interviews

[0334] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). Notification is sent via email or SMS. The input is the interview date, time, and method, and the output is the notification to the candidate. The user (candidate) takes the interview using a video conferencing tool at the notified date and time. During the interview, the device uses a speech generation AI (e.g., Amazon Polly) to read out pre-generated questions. The input is the generated question set, and the output is the audio reading.

[0335] Step 4:

[0336] Analysis by emotion engine

[0337] During the interview, the device runs an emotion recognition engine to analyze the user's (candidate's) facial expressions and tone of voice. Emotional data is extracted in real time and sent to the server. The input is video and audio data, and the output is emotion data. Specifically, a webcam and microphone are used to capture face and voice, and the emotion recognition engine (e.g., Microsoft Azure Emotion API) analyzes this data in real time.

[0338] Step 5:

[0339] Analysis of response and sentiment data

[0340] The server converts the candidate's voice response into text data using speech recognition technology (e.g., Google Speech-to-Text). It then analyzes the response using natural language processing technology. The input is the voice response and emotional data, and the output is the analyzed text and emotional evaluation. The specificity and logic of the response are evaluated, and data obtained from the emotion recognition engine is simultaneously analyzed to evaluate emotional changes. The text data is evaluated using an NLP engine (e.g., BERT).

[0341] Step 6:

[0342] Knowledge test

[0343] The server uses a generative AI model to create knowledge test questions and presents them to the candidate via their terminal. For example, a prompt such as "Please create questions about data structures and algorithms" is input to the generative AI model. The input is the prompt to the generative AI model, and the output is the knowledge test questions. The user answers the test, and the answers are sent back to the server. The input is the candidate's test answers, and the output is the evaluated answers.

[0344] Step 7:

[0345] Creating and Providing Feedback

[0346] The server integrates the results of the interview and knowledge test and creates feedback materials that include the candidate's strengths and weaknesses and what to check next time. The feedback also includes emotional data. The inputs are the interview results, knowledge test results, and emotional data, and the output is the feedback materials. The created feedback materials are sent to the first interviewer. Specifically, a report generation tool (e.g., Tableau) is used to provide feedback in a visually easy-to-understand format.

[0347] (Application example 2)

[0348] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0349] Traditional hiring processes and customer service in brick-and-mortar stores are prone to errors, time-consuming manual operations, and subjective evaluation biases by interviewers and store clerks. Furthermore, traditional systems are unable to recognize users' emotions in real time and respond accordingly. This creates a need for improved hiring efficiency and customer satisfaction.

[0350] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0351] In this invention, the server includes means for receiving and analyzing a resume, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing engine and reading out questions, means for converting voice responses to text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for transmitting the feedback materials to the interviewer, an emotion recognition engine for extracting emotion data, means for analyzing the emotion data to evaluate the user's emotional state, means for answering the user's questions in real time, means for using natural language processing technology to understand the content of the user's questions, and means for analyzing and displaying data input to smart glasses or a head-mounted display. This enables efficient and objective evaluation in the recruitment process and customer service in physical stores, ultimately improving user satisfaction.

[0352] A "resume" is a document submitted by an applicant that includes information such as a self-introduction, educational background, work history, skills, and motivation for applying.

[0353] "Analysis" is the process of breaking down data or information to understand its content for a specific purpose.

[0354] "Natural language processing technology" is technology that enables computers to understand, interpret, and generate human language.

[0355] A "section" is a part of a document or data that is divided based on a specific theme or content.

[0356] A "question set" is a collection containing a series of questions created for a specific purpose.

[0357] A "video conferencing engine" is software or a system for visual and audio communication over the Internet.

[0358] A "voice response" is a response or reply given aloud.

[0359] "Convert to text" is the process of converting non-text data, such as audio or images, into character data.

[0360] "Feedback materials" are documents that compile information including the results of the evaluation and areas for improvement next time.

[0361] An "interviewer" is a person responsible for conducting an interview with an applicant and evaluating the results.

[0362] An "emotion recognition engine" is a system that detects and recognizes emotions from facial expressions, tone of voice, body movements, etc.

[0363] "Emotional state" refers to a person's emotional or psychological state at a given time.

[0364] "Smart glasses" are glasses with additional functions, such as displaying information and providing augmented reality functions.

[0365] A "head-mounted display" is a display device worn on the head to display images in front of the eyes.

[0366] System configuration

[0367] This invention is a system that combines artificial intelligence and emotion recognition technology for the early stages of the hiring process and customer service in brick-and-mortar stores. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate process. It can also be applied to real-time customer service in brick-and-mortar stores.

[0368] Resume analysis

[0369] The server receives resumes uploaded by applicants. It uses optical character recognition (OCR) technology to convert the resume file (PDF or Word file) into text data. It then uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Examples of software used here include Tesseract OCR and spaCy.

[0370] Question Set Generation

[0371] The server generates a set of questions based on the extracted information, customized to include content relevant to the position. For example, a generative AI model could be used to generate questions like, "Tell me about your Python programming experience."

[0372] Conducting interviews

[0373] The device notifies the applicant of the date, time, and method of the interview (for example, a Zoom link using a video conferencing engine). The applicant takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions, and the user (applicant) answers them.

[0374] Incorporating an emotion engine

[0375] During the interview, the device runs an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is sent to a server for analysis. For example, OpenCV or EmotionRecognizer is used.

[0376] Analysis of response and sentiment data

[0377] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. Points such as the specificity and logic of the response are also evaluated. Emotional data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed. As a result, if the user's response is highly specific and emotionally confident, it can be evaluated as an "excellent answer."

[0378] Knowledge test

[0379] The server generates knowledge test questions in a predetermined format and presents them to the applicant via the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[0380] Creating and Providing Feedback

[0381] The server integrates the results of the interview and knowledge test to create feedback materials that include the applicant's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[0382] Application in physical stores

[0383] This system can also be applied to customer service in brick-and-mortar stores. For example, when a customer asks a question, a response can be made in real time. An emotion recognition engine analyzes the video input and evaluates customer satisfaction. This makes it possible to respond to customer needs quickly, which is expected to improve customer satisfaction.

[0384] Prompt Sentence Examples

[0385] "When a customer asks, 'Do you have this shirt in size large?', please provide an appropriate response based on your product inventory data."

[0386] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0387] Processing Steps of the Detailed Description

[0388] Step 1:

[0389] The server receives resumes uploaded by applicants. The submitted resumes are in PDF or Word file format and are converted into text data using optical character recognition (OCR) technology. Specifically, Tesseract OCR is used to analyze the characters in the image and generate text data.

[0390] Step 2:

[0391] The server analyzes the text data using natural language processing (NLP) technology to identify each section, such as educational background, work history, skills, motivation, etc. At this stage, the text is grammatically analyzed using, for example, spaCy, to extract the necessary information.

[0392] Step 3:

[0393] The server generates a set of questions based on the extracted information. Using a generative AI model, it generates questions relevant to the position being applied for. For example, specific questions such as "Tell me about your Python programming experience" are generated based on the prompt. This set of questions is output as text data.

[0394] Step 4:

[0395] The device notifies the applicant of the interview date and method (such as a Zoom link using a video conferencing engine). The notification is sent to the applicant in the form of a message or email. The user (applicant) receives the notification.

[0396] Step 5:

[0397] The user (applicant) is interviewed using a video conferencing tool. During the interview, the device uses a speech generation AI to read out pre-generated questions based on the question set generated in Step 3.

[0398] Step 6:

[0399] The user (applicant) answers questions by voice. The device collects the voice responses in real time and sends the voice data to the server.

[0400] Step 7:

[0401] The server converts the user's voice response into text data using voice recognition technology, such as the Google Speech-to-Text API.

[0402] Step 8:

[0403] The server then analyzes the converted text data using natural language processing technology to evaluate the specificity and logic of the answers. At this stage, text analysis is again performed using tools such as spaCy.

[0404] Step 9:

[0405] During the interview, the device runs an emotion recognition engine, analyzes the user's facial expressions and tone of voice, and extracts emotional data in real time. This emotional data is input as image and audio data and analyzed using OpenCV and EmotionRecognizer.

[0406] Step 10:

[0407] The server analyzes the emotional data and evaluates the user's emotional state. Specifically, it evaluates the user's emotional changes over time and quantifies their confidence and nervousness during the interview.

[0408] Step 11:

[0409] The server combines the analysis results of the interview voice responses with the emotional data to create feedback materials, which include the user's strengths and weaknesses and points to check next time. The feedback materials are output in text format.

[0410] Step 12:

[0411] The server sends the created feedback materials to the interviewer, who is then provided with the materials via email or a dedicated portal.

[0412] Step 13:

[0413] The server generates questions for the knowledge test in a predetermined format and presents them to the applicant via the terminal. The questions for the knowledge test are entered in text format, and the user answers them.

[0414] Step 14:

[0415] The user answers the knowledge test questions and sends the answer data to the server, which evaluates the knowledge test answers and reflects the results in the feedback material.

[0416] Step 15:

[0417] In a physical store, the device uses natural language processing technology and generative AI models to answer user questions in real time. For example, if a user asks, "Do you have this shirt in size L?", the device will provide an appropriate answer based on inventory data.

[0418] Step 16:

[0419] The device uses smart glasses or a head-mounted display to analyze the input data and provide visual information to the user. The user's visual data is analyzed in real time, and appropriate product information and guidance are displayed.

[0420] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0421] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0422] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0423] [Second embodiment]

[0424] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0425] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0426] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0427] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0428] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0429] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0430] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0431] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0432] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0433] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0434] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0435] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0436] System Overview

[0437] This invention is a system that uses artificial intelligence (generative AI models) to conduct interviews at the early stage of the recruitment process, reducing recruitment costs and preventing mismatches.

[0438] Resume analysis

[0439] The server receives the resume uploaded by the candidate and converts it into text data using optical character recognition (OCR) technology, then uses natural language processing technology to identify sections of the resume and extract information such as educational background, work history, skills, and motivation for applying.

[0440] Question Set Generation

[0441] Based on the extracted information, the server generates a set of interview questions for the candidate, which are customized according to the company's requirements and the candidate's characteristics.

[0442] Conducting interviews

[0443] The device notifies the candidate of the date, time, and method of the interview. The candidate will be interviewed at the specified date and time using a video conferencing tool (e.g., Zoom). Questions are read aloud by a voice generation AI, and the user (candidate) answers them.

[0444] Analysis of responses

[0445] The server converts the candidate's voice responses into text in real time, which is then analyzed using natural language processing technology to evaluate the specificity and logic of the responses.

[0446] Knowledge test

[0447] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal, and the candidate's answers are sent to the server for evaluation.

[0448] Creating and Providing Feedback

[0449] Based on the results of the interview and knowledge test, the server creates feedback materials, including the candidate's strengths and weaknesses and points to check next time. The created feedback materials are sent to the first interviewer.

[0450] Specific examples of processing

[0451] For example, when a software engineer applicant uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as, "Please tell us about your specific project experience using Python and Django" are generated. On the day of the interview, the voice generation AI on the device reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0452] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0453] The processing flow will be explained below.

[0454] Step 1:

[0455] The server receives resume files uploaded by applicants, which are assumed to be in PDF or Word format.

[0456] Step 2:

[0457] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[0458] Step 3:

[0459] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Specifically, it uses a text analysis algorithm.

[0460] Step 4:

[0461] The server generates a set of questions based on the extracted information, taking into account questions relevant to the position, such as "Tell me about your Python programming experience."

[0462] Step 5:

[0463] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link). Notifications will be sent via email or messaging system.

[0464] Step 6:

[0465] Users are interviewed at the appointed time using a video conferencing tool (such as Zoom), and the device reads out pre-generated questions using speech generation AI.

[0466] Step 7:

[0467] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[0468] Step 8:

[0469] The server converts the user's response voice into text data using speech recognition technology, and the converted text data is analyzed using natural language processing technology.

[0470] Step 9:

[0471] The server evaluates the analyzed answer text and assigns a score based on points such as specificity and logic.

[0472] Step 10:

[0473] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[0474] Step 11:

[0475] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[0476] Step 12:

[0477] The server evaluates the candidate's knowledge test answers and calculates a score.

[0478] Step 13:

[0479] The server combines the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what needs to be checked next time.

[0480] Step 14:

[0481] The server sends the created feedback material to the first interviewer, who uses the material to prepare for the next interview.

[0482] These are the specific steps for efficiently conducting the recruitment process. This system reduces recruitment costs and enables the selection of appropriate personnel.

[0483] Example 1

[0484] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0485] A problem with the traditional recruitment process is that the initial interview stage takes a lot of time and costs a lot of money. Furthermore, evaluations often depend on the interviewer's subjective judgment, making it difficult to objectively evaluate the candidate's aptitude and abilities. For these reasons, there is a need to prevent mismatches and streamline the recruitment process.

[0486] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0487] In this invention, the server includes means for receiving and analyzing resumes, means for converting resumes into text data using optical character recognition technology, means for identifying each section of the resume and extracting information using natural language processing technology, means for generating a set of interview questions based on the extracted information, means for customizing the questions according to the company's requirements and the applicant's characteristics, means for notifying the applicant of the interview date and method, means for conducting the interview using a video conferencing tool and reading the questions aloud using speech generation technology, means for converting the user's voice responses into text, means for analyzing and evaluating the converted text using natural language processing technology, means for creating feedback materials based on the results of the interview and knowledge test, and means for sending the feedback materials to the interviewer, thereby enabling an efficient hiring process and objective evaluation.

[0488] "Means for receiving and analyzing resumes" refers to a series of functions for receiving resumes from applicants and analyzing their contents.

[0489] "Optical character recognition technology" is a technology that extracts character information from documents such as images and PDFs.

[0490] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[0491] "Means for identifying each section of a resume and extracting information" refers to a function that analyzes the contents of a resume to identify each section, such as educational background, work history, and skills, and extracts that information.

[0492] The "means for generating a set of interview questions" is a function that creates questions for candidates based on the analyzed resume information.

[0493] "A means of customizing questions according to the company's requirements and the characteristics of the applicant" refers to a function that appropriately changes questions based on the skills the company is looking for and the background of the applicant.

[0494] "Means of notifying applicants of the interview date and method" is a function that informs applicants of the date and time of the interview and how it will be conducted.

[0495] "Means for conducting interviews using video conferencing tools and reading questions aloud using voice generation technology" refers to a function that conducts interviews using video conferencing software and presents generated questions to candidates as audio.

[0496] The "means for converting the user's voice response into text" is a function for converting the content of the user's verbal response into text data in real time.

[0497] "Means for analyzing and evaluating the converted text using natural language processing technology" refers to a function that analyzes text converted from speech and evaluates the specificity and logic of the content.

[0498] The "means for creating feedback materials" is a function for creating feedback documents for applicants based on the result data of the interview and knowledge test.

[0499] The "means of sending feedback materials to interviewers" is a function for sending the created feedback materials to the first interviewer and other related staff.

[0500] This invention relates to a system that uses artificial intelligence (generative AI models) to conduct interviews in the early stages of the recruitment process, which reduces recruitment costs and prevents mismatches.

[0501] System Overview

[0502] The system is designed to automate the analysis of resumes, generation of question sets, conducting interviews, analyzing responses, conducting knowledge tests, and creating and providing feedback. Each function is described in detail below.

[0503] Resume analysis

[0504] 1. A user accesses the system and uploads a resume.

[0505] 2. The server receives the resume and converts it into text data using optical character recognition (OCR) technology. Specifically, it uses Tesseract OCR.

[0506] 3. The server analyzes the converted text data using natural language processing technology (e.g., spaCy), identifies each section of the resume (educational history, work history, skills, motivation, etc.), and extracts the necessary information.

[0507] Question Set Generation

[0508] 1. The server generates a set of interview questions based on the extracted information, customizing the questions according to the company's requirements and the applicant's characteristics.

[0509] 2. For example, the server uses a generative AI model such as GPT-3 to create questions such as, "Tell me about your specific project experience using Python and Django."

[0510] Conducting interviews

[0511] 1. The device notifies candidates of the date, time, and method of their interview via email, SMS, or a dedicated app.

[0512] 2. Candidates will be interviewed via video conferencing tools (e.g., Zoom) at the appointed date and time.

[0513] 3. During the interview, a voice generation AI (e.g., Google Text-to-Speech) on the device reads out the questions retrieved from the server.

[0514] Analysis of responses

[0515] 1. The candidate answers and the device records the audio.

[0516] 2. The server converts the recorded audio into text in real time using Google Speech-to-Text.

[0517] 3. The converted text is analyzed using natural language processing technology (e.g., spaCy) to evaluate the specificity and logic of the answers.

[0518] Knowledge test

[0519] 1. The server generates knowledge test questions and presents them to the candidate in a predetermined format. The test content includes coding questions and theory questions.

[0520] 2. The candidate answers, and the answer is sent to the server for evaluation.

[0521] Creating and Providing Feedback

[0522] 1. The server creates feedback materials based on the results of the interview and knowledge test. The feedback materials include the candidate's strengths and weaknesses and what needs to be checked next time.

[0523] 2. The prepared feedback materials will be sent to the first interviewer.

[0524] Specific examples

[0525] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as "Tell us about your specific project experience using Python and Django" are generated. On the day of the interview, a voice generation AI reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0526] Example prompt sentence:

[0527] "Design a system that parses resumes uploaded by candidates, generates interview questions, conducts interviews, parses responses, administers knowledge tests, and generates feedback materials."

[0528] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0529] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0530] Step 1:

[0531] A user accesses the system and uploads a resume. The input is the user's resume file, and the output is the resume data stored on the server. Specifically, the user creates a profile through the web interface and clicks the resume upload button. The server receives this upload request and saves the resume.

[0532] Step 2:

[0533] The server receives the resume and converts it into text data using optical character recognition (OCR) technology. The input is the uploaded resume file, and the output is the converted text data. Specifically, the server invokes Tesseract OCR to extract text from the resume image.

[0534] Step 3:

[0535] The server uses natural language processing technology to analyze resumes and extract the necessary information. The input is text data, and the output is extracted information such as educational background, work history, skills, and motivation for applying. The server uses spaCy and NLTK to analyze the text and identify keywords and sections.

[0536] Step 4:

[0537] The server generates a set of interview questions based on the extracted information. The input is the extracted information, and the output is the generated set of questions. Specifically, the server retrieves requirements from the company's database and creates questions using a generative AI model (e.g., GPT-3).

[0538] Step 5:

[0539] The server customizes the questions according to the company's requirements and the applicant's characteristics. The input is the generated question set and the company's requirements, and the output is the customized question set. Specifically, the server performs filtering and text fine-tuning on the generated questions.

[0540] Step 6:

[0541] The device notifies the candidate of the interview date, time, and method. The input is information about the interview date, time, and method, and the output is the notified candidate. The device sends this information to the candidate via email, SMS, or a dedicated app.

[0542] Step 7:

[0543] A candidate will be interviewed at a specified date and time using a video conferencing tool. The input is the interview link and date and time information, and the output is joining the video conference. Specifically, the user will log in to the video conferencing tool (e.g., Zoom) using their device.

[0544] Step 8:

[0545] The device uses speech generation AI to read out questions. The input is a customized set of questions, and the output is audio questions. Specifically, the device uses the Google Text-to-Speech API to convert the questions into audio and present them to the candidate as audio played through the speaker.

[0546] Step 9:

[0547] The user (candidate) answers questions, and the device records the audio. The input is the user's voice response, and the output is the recorded audio data. Specifically, the device captures the audio using Zoom's recording function or a dedicated app.

[0548] Step 10:

[0549] The server converts the user's voice response into text. The input is the recorded voice data, and the output is the converted text data. Specifically, the server calls Google Speech-to-Text to convert the voice to text.

[0550] Step 11:

[0551] The server analyzes the converted text and evaluates the specificity and logic of the answer. The input is text data, and the output is the analysis result. Specifically, the server uses spaCy to analyze the text and evaluates it based on specific criteria.

[0552] Step 12:

[0553] The server generates knowledge test questions and presents them to candidates through their terminals. The input is the knowledge test requirements, and the output is a set of knowledge test questions. Specifically, the server uses a question generation algorithm to create coding questions and theory questions and display them on the terminal.

[0554] Step 13:

[0555] The candidate's answers are sent through the terminal and evaluated by the server. The input is the candidate's answer data, and the output is the evaluation result. Specifically, the terminal sends the answer data to the server, which then automatically scores it.

[0556] Step 14:

[0557] The server creates feedback documents based on the results of the interview and knowledge test. The input is the result data of the interview and knowledge test, and the output is the feedback documents. Specifically, the server aggregates the results and generates the feedback documents using a generative AI model.

[0558] Step 15:

[0559] The created feedback material is sent to the first interviewer. The input is the feedback material, and the output is the feedback sent to the first interviewer. In concrete terms, the server sends the created feedback material to the first interviewer via email or an internal system.

[0560] (Application example 1)

[0561] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0562] Maintenance work for factory robots is often complex and requires specialized knowledge, making the judgment of maintenance workers essential. However, because maintenance depends on the skills and experience of the workers, there is a problem of inconsistency in the quality of maintenance. Another issue is the lack of a means to efficiently provide work procedures and provide appropriate feedback in real time, which reduces work efficiency.

[0563] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0564] In this invention, the server includes means for receiving and analyzing a resume, means for identifying each section of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing tool and reading out the questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for sending the feedback materials to the interviewer, means for extracting text information from an image using optical character recognition technology, means for generating a maintenance procedure tailored to a specific situation using a generative artificial intelligence model, and means for evaluating the maintenance work based on the procedure, thereby enabling the efficiency and quality of robot maintenance work in a factory to be improved.

[0565] "Means for receiving and analyzing resumes" refers to a function for electronically receiving resumes submitted by applicants and analyzing their contents.

[0566] "Means for identifying each section of a resume using natural language processing technology" is a function that uses natural language processing technology to classify and identify the contents of a resume into sections such as educational history, work history, and skills.

[0567] The "means for generating a question set based on the analysis results" is a function for generating appropriate interview questions using the analysis results of a resume.

[0568] The "means for conducting an interview using a video conferencing tool and reading out questions" is a function for conducting an interview using a video conferencing tool and reading out the generated questions aloud.

[0569] "Means for converting voice responses into text and analyzing it using natural language processing technology" refers to a function for converting the applicant's voice responses into text data and analyzing that text using natural language processing technology.

[0570] The "means for evaluating the analysis results and creating feedback materials" is a function for evaluating the analysis results of the voice responses and creating materials summarizing feedback to the applicant.

[0571] The "means for sending feedback materials to interviewers" is a function for electronically sending the created feedback materials to interviewers.

[0572] "Means for extracting character information from an image using optical character recognition technology" is a function for reading character information in an image using optical character recognition technology and converting it into text data.

[0573] "Means for generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model" is a function for automatically generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model.

[0574] The "means for evaluating maintenance work based on the procedure" is a function for evaluating work performed based on the generated maintenance procedure.

[0575] This invention relates to a system for improving the efficiency and quality of robot maintenance work in a factory. In one embodiment of the present invention, how a server, a terminal, and a user work together to realize the system will be described.

[0576] Server processing

[0577] Resume analysis

[0578] The server receives the resume uploaded by the user (worker) and converts it into text data using optical character recognition (OCR) technology. The specific software used is OpenCV and pytesseract. After this, each section of the resume (educational history, work history, skills, etc.) is identified using natural language processing technology. The specific software used in this step is Python and a library for natural language processing.

[0579] Question Set Generation

[0580] The server generates a set of interview questions based on the analysis results. This part uses a generative AI model (e.g., GPT-3.5) to create questions that match the characteristics of the candidate and the company's requirements. These questions will also be used to generate maintenance procedures in the future.

[0581] Processing by the terminal

[0582] Conducting interviews

[0583] The device notifies the user of the date, time, and method of the interview. The user then attends the interview at the specified time using a video conferencing tool (e.g., Zoom). The questions are read aloud by a speech generation AI. This part uses Google Cloud Speech-to-Text and Text-to-Speech services for speech recognition and generation technology.

[0584] Analysis of responses

[0585] The server converts the user's voice response into text in real time using Google Cloud Speech-to-Text, and the converted text is analyzed using natural language processing technology to evaluate the specificity and logic of the response.

[0586] Robot Maintenance

[0587] Extracting character information using optical character recognition technology

[0588] The server extracts the robot's state information from the images using optical character recognition techniques, specifically OpenCV and pytesseract.

[0589] Generate maintenance procedures

[0590] The server uses a generative AI model to generate maintenance procedures for specific situations. GPT-3.5 is used to generate these procedures. For example, if the robot's status image displays "Error message 72: Motor overheating," the following prompt sentence is input to the generative AI model:

[0591] The robot status is displayed as "Error message 72: Motor overheated". Please suggest the following maintenance steps:

[0592] Maintenance work evaluation

[0593] The server evaluates the work performed based on the generated maintenance procedure and creates feedback materials based on the evaluation results, which are provided to the workers and used to improve future maintenance work.

[0594] As a result, the efficiency and quality of robot maintenance work within the factory will be improved.

[0595] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0596] Step 1:

[0597] The server receives resumes uploaded by users. The input is the resume uploaded by the user in PDF or image format. The server converts this resume into text data using optical character recognition (OCR) technology. Specifically, it uses OpenCV and pytesseract to recognize characters in the image and extract them as text data. The output is the text data of the resume.

[0598] Step 2:

[0599] The server analyzes the text data extracted by OCR using natural language processing technology. The input is the text data of a resume. The server analyzes this data and identifies each section, such as educational history, work history, and skills. Specifically, it uses Python and natural language processing libraries. The output is the text data of a resume with each section identified.

[0600] Step 3:

[0601] The server generates a set of interview questions based on the analysis results. The input is the text data for each identified section. The server uses a generative AI model (e.g., GPT-3.5) to create a question set that corresponds to the candidate's characteristics and the company's requirements. The output is the generated question set.

[0602] Step 4:

[0603] The terminal notifies the user of the date, time, and method of the interview. The input is the question set and interview schedule data generated from the server. The terminal sends the notification to the user using email or a messaging service. The output is data about the date, time, and method of the notification.

[0604] Step 5:

[0605] The user is interviewed at a specified date and time using a video conferencing tool (e.g., Zoom). The input is a set of questions generated by the server and the user's answers. A voice generation AI on the device reads the questions aloud. The output is the data of the user's voice responses.

[0606] Step 6:

[0607] The server converts the user's voice response into text data in real time. The input is the user's voice response. The server converts the voice to text data using Google Cloud Speech-to-Text technology. The output is the text data response.

[0608] Step 7:

[0609] The server analyzes the text data responses using natural language processing technology. The input is the text data responses. The server analyzes the converted text and evaluates the specificity and logic of the responses. The output is evaluation data of the analysis results.

[0610] Step 8:

[0611] The server creates feedback materials based on the analysis results. The input is the evaluation data. The server generates feedback sentences using a Python script. The output is the feedback materials.

[0612] Step 9:

[0613] The server sends the created feedback materials to the interviewer. The input is the feedback materials. The server sends the materials via email. The output is a notification that the materials have been sent.

[0614] Step 10:

[0615] The server uses optical character recognition technology to extract robot status information from images. The input is image data showing the robot's status. The server uses OpenCV and pytesseract to convert the character information in the image into text data. The output is text data of the robot's status.

[0616] Step 11:

[0617] The server uses a generative artificial intelligence model to generate maintenance procedures tailored to specific situations. The input is text data of the robot's status. The server uses a generative AI model (e.g., GPT-3.5) to generate appropriate maintenance procedures. For example, the prompt sentence "The robot's status is displayed as 'Error message 72: Motor overheating'. Please suggest the next maintenance procedure." is input to the model. The output is the generated maintenance procedure.

[0618] Step 12:

[0619] The server evaluates the maintenance work based on the procedure. The input is the generated maintenance procedure and the maintenance work result data. The server evaluates the work performed according to the procedure and creates the results as feedback material. The output is the evaluation result of the maintenance work and the feedback material.

[0620] Step 13:

[0621] The server sends the created feedback material to the worker. The input is the feedback material. The server sends the material via email. The output is a notification that the material has been sent.

[0622] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0623] System Overview

[0624] This invention is a system that uses artificial intelligence (generative AI models) in the early stages of the hiring process. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate hiring process.

[0625] Resume analysis

[0626] The server receives resumes uploaded by candidates, converts the resume files (PDF or Word files) into text data using optical character recognition (OCR), and then uses natural language processing (NLP) to extract information such as educational background, work history, skills, and motivation for applying from the resume text.

[0627] Question Set Generation

[0628] The server generates a set of questions based on the extracted information, customized to include content relevant to the position being applied for. For example, a software engineer candidate might be asked questions like, "Tell me about your Python programming experience."

[0629] Conducting interviews

[0630] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). The candidate takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions. The user (candidate) answers them.

[0631] Incorporating an emotion engine

[0632] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is then sent to the server.

[0633] Analysis of response and sentiment data

[0634] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. The response is evaluated based on factors such as specificity and logic. Emotion data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed.

[0635] Knowledge test

[0636] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[0637] Creating and Providing Feedback

[0638] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[0639] Specific examples of processing

[0640] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. During the interview, they are asked questions such as, "Tell us about your specific project experience using Python and Django," and the emotion engine analyzes the user's facial expressions and tone of voice while answering. The server analyzes the voice responses and emotion data and makes an evaluation such as, "The answers were highly specific and emotionally confident." This evaluation is reflected in the feedback materials and sent to the first interviewer.

[0641] By incorporating emotion recognition technology, this system enables objective evaluation that is not influenced by the interviewer's subjectivity, contributing to the selection of more appropriate personnel. It also improves convenience for applicants by providing flexible interview opportunities.

[0642] The processing flow will be explained below.

[0643] Step 1:

[0644] The server receives resume files uploaded by applicants, which are expected to be in PDF or Word format.

[0645] Step 2:

[0646] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[0647] Step 3:

[0648] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data.

[0649] Step 4:

[0650] The server generates a set of questions based on the extracted information, including questions such as "Tell me about your Python programming experience."

[0651] Step 5:

[0652] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link) via email or messaging system.

[0653] Step 6:

[0654] The user will be interviewed at the appointed date and time using a video conferencing tool (Zoom), with the device reading out pre-generated questions using a speech-generating AI.

[0655] Step 7:

[0656] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[0657] Step 8:

[0658] During the interview, the device activates an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time, which is then sent to the server.

[0659] Step 9:

[0660] The server uses speech recognition technology to convert the user's voice response into text data. The converted text is then analyzed using natural language processing technology to evaluate points such as the specificity and logic of the response. At the same time, emotional data obtained from an emotion recognition engine is also analyzed.

[0661] Step 10:

[0662] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[0663] Step 11:

[0664] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[0665] Step 12:

[0666] The server evaluates the candidate's knowledge test answers and calculates a score.

[0667] Step 13:

[0668] The server aggregates all responses and emotional data to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. The emotional data indicates the candidate's emotional changes during the interview and their overall emotional state.

[0669] Step 14:

[0670] The server transmits the created feedback material to the first interviewer, who can use this material to efficiently prepare for the next interview.

[0671] Example 2

[0672] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0673] In the traditional hiring process, it is easily influenced by the interviewer's subjectivity, making it difficult to make an objective evaluation. Furthermore, there is no way to evaluate the candidate's emotional changes and facial expressions in real time, making it difficult to make an accurate and fair evaluation. Furthermore, the typical interview process is time-consuming, labor-intensive, and inefficient.

[0674] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0675] In this invention, the server includes means for receiving and analyzing resumes, means for converting resume files into text data using optical character recognition technology, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for inputting prompts and generating appropriate questions using a generative AI model, means for conducting interviews using a video conferencing tool and reading out questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for analyzing facial expressions and tone of voice during the interview using an emotion recognition engine and extracting emotion data, means for integrating the analysis results and the emotion data and evaluating them, means for creating feedback materials, and means for sending the feedback materials to interviewers. This enables objective and accurate evaluation, including changes in the emotions of candidates, and improves the efficiency and fairness of the hiring process.

[0676] "Means for receiving and analyzing resumes" refers to the device or software used to receive resume files uploaded by applicants and analyze their contents.

[0677] "Optical character recognition technology" is a technology that extracts text information from image data and is used to convert digital resume files into text data.

[0678] "Natural language processing technology" is a technology that uses computers to analyze and understand text written in natural language, and is used to extract necessary information from text data such as resumes.

[0679] A "means for generating a question set" is a device or software for generating appropriate interview questions based on the analyzed resume data.

[0680] A "generative AI model" is a trained artificial intelligence model that has the ability to generate natural language when given a prompt sentence.

[0681] A "prompt" is an instruction given to an AI model to perform a specific task.

[0682] A "video conferencing tool" is software for sharing audio and video in real time over the Internet.

[0683] A "means for converting voice responses to text" is a device or software that uses voice recognition technology to convert voice responses during an interview into text form.

[0684] An "emotion recognition engine" is a device or software that analyzes a person's emotional state from audio and video data and extracts emotional data in real time.

[0685] The "means for creating feedback materials" refers to a device or software for generating feedback materials summarizing the evaluation of candidates based on the results of the interview and knowledge test.

[0686] The "means for sending feedback materials to interviewers" refers to a device or software for automatically sending the created feedback materials to interviewers.

[0687] This invention is a system that combines artificial intelligence (generative AI models) and an emotion recognition engine to perform more accurate evaluations in the early stages of the hiring process. This system automates the processes of resume analysis, question generation, interviews, response analysis, knowledge testing, and feedback creation.

[0688] Hardware and software used

[0689] The system uses the following hardware and software:

[0690] 1. Server

[0691] A server with the storage and processing power to receive resumes

[0692] Optical Character Recognition (OCR) engine (e.g., Tesseract)

[0693] Natural Language Processing (NLP) engines (e.g., SpaCy, BERT)

[0694] Generative AI models (e.g., GPT-3)

[0695] Voice recognition technology (e.g., Google Speech-to-Text)

[0696] Report generation tools (e.g. Tableau)

[0697] 2. Terminal

[0698] Devices for real-time communication with users (PCs, smartphones, tablets, etc.)

[0699] Video conferencing tools (e.g., Zoom, Microsoft Teams)

[0700] Voice generation AI (e.g. Amazon Polly)

[0701] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[0702] 3. Users

[0703] Applicants and interviewers

[0704] Specific processing of the program

[0705] Resume upload and analysis

[0706] Users upload their resumes to the server, which converts the received resume files (PDF or Word files) into text data using an optical character recognition (OCR) engine. It then uses natural language processing (NLP) to extract information such as educational background, work history, and skills from the resume text data.

[0707] Question Set Generation

[0708] The server generates a set of relevant questions for the candidate based on the analyzed resume data. It uses a generative AI model (GPT-3) to generate appropriate questions. For example, the server inputs a prompt such as, "Generate interview questions based on the candidate's programming skills."

[0709] Preparing for and conducting interviews

[0710] The device notifies the candidate of the date, time, and method of the interview, sends a link to a video conferencing tool (such as Zoom), and the interview takes place at the specified date and time. During the interview, a speech-generation AI (Amazon Polly) is used to read out pre-generated questions.

[0711] Analysis by emotion engine

[0712] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice, extracting emotional data in real time and sending it to the server.

[0713] Analysis of response and sentiment data

[0714] The server converts the applicant's voice responses into text data using speech recognition technology and analyzes them using natural language processing technology. The server evaluates the specificity and logic of the responses. It also analyzes emotional data obtained from an emotion recognition engine to evaluate emotional changes.

[0715] Knowledge test

[0716] The server creates knowledge test questions using a generative AI model and presents them to the candidate via their device. For example, the candidate might enter a prompt such as, "Please create a question about data structures and algorithms." The candidate's answers are then sent to the server for evaluation.

[0717] Creating and Providing Feedback

[0718] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. This feedback includes emotional data and is evaluated in detail. The created feedback materials are sent to the first interviewer.

[0719] Examples of prompt statements

[0720] 1. "Please parse this resume and extract Python and Django project experience."

[0721] 2. "Generate questions to assess knowledge of data structures and algorithms."

[0722] 3. "Evaluate the specificity of interview responses and create feedback materials that analyze emotional data."

[0723] By integrating emotion recognition and AI technology, the system eliminates subjective interviewer evaluations, significantly improving the accuracy and efficiency of the hiring process, while providing candidates with a flexible interview experience for increased convenience.

[0724] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0725] Step 1:

[0726] Resume upload and analysis

[0727] The user uploads their resume to the system. The server stores the received resume file (PDF or Word file) in storage. Optical character recognition (OCR) technology is used to convert the file into text data. The input is a PDF or Word file, and the output is text data. The server launches an OCR engine (e.g., Tesseract) to analyze the file and extract the text. Next, natural language processing (NLP) technology is used to extract information such as education, work history, and skills from the resume text data. An NLP engine (e.g., SpaCy) is used to identify the information for each section.

[0728] Step 2:

[0729] Question Set Generation

[0730] The server generates a set of questions relevant to the candidate based on the analyzed information. It uses a generative AI model (e.g., GPT-3) and inputs a prompt to generate questions. The input is the analyzed resume data and the prompt, and the output is a set of interview questions. For example, the prompt "Generate interview questions based on the candidate's programming skills" is input to the generative AI model. The generated questions are saved for interview preparation.

[0731] Step 3:

[0732] Preparing for and conducting interviews

[0733] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). Notification is sent via email or SMS. The input is the interview date, time, and method, and the output is the notification to the candidate. The user (candidate) takes the interview using a video conferencing tool at the notified date and time. During the interview, the device uses a speech generation AI (e.g., Amazon Polly) to read out pre-generated questions. The input is the generated question set, and the output is the audio reading.

[0734] Step 4:

[0735] Analysis by emotion engine

[0736] During the interview, the device runs an emotion recognition engine to analyze the user's (candidate's) facial expressions and tone of voice. Emotional data is extracted in real time and sent to the server. The input is video and audio data, and the output is emotion data. Specifically, a webcam and microphone are used to capture face and voice, and the emotion recognition engine (e.g., Microsoft Azure Emotion API) analyzes this data in real time.

[0737] Step 5:

[0738] Analysis of response and sentiment data

[0739] The server converts the candidate's voice response into text data using speech recognition technology (e.g., Google Speech-to-Text). It then analyzes the response using natural language processing technology. The input is the voice response and emotional data, and the output is the analyzed text and emotional evaluation. The specificity and logic of the response are evaluated, and data obtained from the emotion recognition engine is simultaneously analyzed to evaluate emotional changes. The text data is evaluated using an NLP engine (e.g., BERT).

[0740] Step 6:

[0741] Knowledge test

[0742] The server uses a generative AI model to create knowledge test questions and presents them to the candidate via their terminal. For example, a prompt such as "Please create questions about data structures and algorithms" is input to the generative AI model. The input is the prompt to the generative AI model, and the output is the knowledge test questions. The user answers the test, and the answers are sent back to the server. The input is the candidate's test answers, and the output is the evaluated answers.

[0743] Step 7:

[0744] Creating and Providing Feedback

[0745] The server integrates the results of the interview and knowledge test and creates feedback materials that include the candidate's strengths and weaknesses and what to check next time. The feedback also includes emotional data. The inputs are the interview results, knowledge test results, and emotional data, and the output is the feedback materials. The created feedback materials are sent to the first interviewer. Specifically, a report generation tool (e.g., Tableau) is used to provide feedback in a visually easy-to-understand format.

[0746] (Application example 2)

[0747] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0748] Traditional hiring processes and customer service in brick-and-mortar stores are prone to errors, time-consuming manual operations, and subjective evaluation biases by interviewers and store clerks. Furthermore, traditional systems are unable to recognize users' emotions in real time and respond accordingly. This creates a need for improved hiring efficiency and customer satisfaction.

[0749] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0750] In this invention, the server includes means for receiving and analyzing a resume, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing engine and reading out questions, means for converting voice responses to text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for transmitting the feedback materials to the interviewer, an emotion recognition engine for extracting emotion data, means for analyzing the emotion data to evaluate the user's emotional state, means for answering the user's questions in real time, means for using natural language processing technology to understand the content of the user's questions, and means for analyzing and displaying data input to smart glasses or a head-mounted display. This enables efficient and objective evaluation in the recruitment process and customer service in physical stores, ultimately improving user satisfaction.

[0751] A "resume" is a document submitted by an applicant that includes information such as a self-introduction, educational background, work history, skills, and motivation for applying.

[0752] "Analysis" is the process of breaking down data or information to understand its content for a specific purpose.

[0753] "Natural language processing technology" is technology that enables computers to understand, interpret, and generate human language.

[0754] A "section" is a part of a document or data that is divided based on a specific theme or content.

[0755] A "question set" is a collection containing a series of questions created for a specific purpose.

[0756] A "video conferencing engine" is software or a system for visual and audio communication over the Internet.

[0757] A "voice response" is a response or reply given aloud.

[0758] "Convert to text" is the process of converting non-text data, such as audio or images, into character data.

[0759] "Feedback materials" are documents that compile information including the results of the evaluation and areas for improvement next time.

[0760] An "interviewer" is a person responsible for conducting an interview with an applicant and evaluating the results.

[0761] An "emotion recognition engine" is a system that detects and recognizes emotions from facial expressions, tone of voice, body movements, etc.

[0762] "Emotional state" refers to a person's emotional or psychological state at a given time.

[0763] "Smart glasses" are glasses with additional functions, such as displaying information and providing augmented reality functions.

[0764] A "head-mounted display" is a display device worn on the head to display images in front of the eyes.

[0765] System configuration

[0766] This invention is a system that combines artificial intelligence and emotion recognition technology for the early stages of the hiring process and customer service in brick-and-mortar stores. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate process. It can also be applied to real-time customer service in brick-and-mortar stores.

[0767] Resume analysis

[0768] The server receives resumes uploaded by applicants. It uses optical character recognition (OCR) technology to convert the resume file (PDF or Word file) into text data. It then uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Examples of software used here include Tesseract OCR and spaCy.

[0769] Question Set Generation

[0770] The server generates a set of questions based on the extracted information, customized to include content relevant to the position. For example, a generative AI model could be used to generate questions like, "Tell me about your Python programming experience."

[0771] Conducting interviews

[0772] The device notifies the applicant of the date, time, and method of the interview (for example, a Zoom link using a video conferencing engine). The applicant takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions, and the user (applicant) answers them.

[0773] Incorporating an emotion engine

[0774] During the interview, the device runs an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is sent to a server for analysis. For example, OpenCV or EmotionRecognizer is used.

[0775] Analysis of response and sentiment data

[0776] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. Points such as the specificity and logic of the response are also evaluated. Emotional data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed. As a result, if the user's response is highly specific and emotionally confident, it can be evaluated as an "excellent answer."

[0777] Knowledge test

[0778] The server generates knowledge test questions in a predetermined format and presents them to the applicant via the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[0779] Creating and Providing Feedback

[0780] The server integrates the results of the interview and knowledge test to create feedback materials that include the applicant's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[0781] Application in physical stores

[0782] This system can also be applied to customer service in brick-and-mortar stores. For example, when a customer asks a question, a response can be made in real time. An emotion recognition engine analyzes the video input and evaluates customer satisfaction. This makes it possible to respond to customer needs quickly, which is expected to improve customer satisfaction.

[0783] Prompt Sentence Examples

[0784] "When a customer asks, 'Do you have this shirt in size large?', please provide an appropriate response based on your product inventory data."

[0785] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0786] Processing Steps of the Detailed Description

[0787] Step 1:

[0788] The server receives resumes uploaded by applicants. The submitted resumes are in PDF or Word file format and are converted into text data using optical character recognition (OCR) technology. Specifically, Tesseract OCR is used to analyze the characters in the image and generate text data.

[0789] Step 2:

[0790] The server analyzes the text data using natural language processing (NLP) technology to identify each section, such as educational background, work history, skills, motivation, etc. At this stage, the text is grammatically analyzed using, for example, spaCy, to extract the necessary information.

[0791] Step 3:

[0792] The server generates a set of questions based on the extracted information. Using a generative AI model, it generates questions relevant to the position being applied for. For example, specific questions such as "Tell me about your Python programming experience" are generated based on the prompt. This set of questions is output as text data.

[0793] Step 4:

[0794] The device notifies the applicant of the interview date and method (such as a Zoom link using a video conferencing engine). The notification is sent to the applicant in the form of a message or email. The user (applicant) receives the notification.

[0795] Step 5:

[0796] The user (applicant) is interviewed using a video conferencing tool. During the interview, the device uses a speech generation AI to read out pre-generated questions based on the question set generated in Step 3.

[0797] Step 6:

[0798] The user (applicant) answers questions by voice. The device collects the voice responses in real time and sends the voice data to the server.

[0799] Step 7:

[0800] The server converts the user's voice response into text data using voice recognition technology, such as the Google Speech-to-Text API.

[0801] Step 8:

[0802] The server then analyzes the converted text data using natural language processing technology to evaluate the specificity and logic of the answers. At this stage, text analysis is again performed using tools such as spaCy.

[0803] Step 9:

[0804] During the interview, the device runs an emotion recognition engine, analyzes the user's facial expressions and tone of voice, and extracts emotional data in real time. This emotional data is input as image and audio data and analyzed using OpenCV and EmotionRecognizer.

[0805] Step 10:

[0806] The server analyzes the emotional data and evaluates the user's emotional state. Specifically, it evaluates the user's emotional changes over time and quantifies their confidence and nervousness during the interview.

[0807] Step 11:

[0808] The server combines the analysis results of the interview voice responses with the emotional data to create feedback materials, which include the user's strengths and weaknesses and points to check next time. The feedback materials are output in text format.

[0809] Step 12:

[0810] The server sends the created feedback materials to the interviewer, who is then provided with the materials via email or a dedicated portal.

[0811] Step 13:

[0812] The server generates questions for the knowledge test in a predetermined format and presents them to the applicant via the terminal. The questions for the knowledge test are entered in text format, and the user answers them.

[0813] Step 14:

[0814] The user answers the knowledge test questions and sends the answer data to the server, which evaluates the knowledge test answers and reflects the results in the feedback material.

[0815] Step 15:

[0816] In a physical store, the device uses natural language processing technology and generative AI models to answer user questions in real time. For example, if a user asks, "Do you have this shirt in size L?", the device will provide an appropriate answer based on inventory data.

[0817] Step 16:

[0818] The device uses smart glasses or a head-mounted display to analyze the input data and provide visual information to the user. The user's visual data is analyzed in real time, and appropriate product information and guidance are displayed.

[0819] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0820] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0821] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0822] [Third embodiment]

[0823] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0824] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0825] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0826] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0827] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0828] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0829] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0830] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0831] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0832] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0833] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0834] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0835] System Overview

[0836] This invention is a system that uses artificial intelligence (generative AI models) to conduct interviews at the early stage of the recruitment process, reducing recruitment costs and preventing mismatches.

[0837] Resume analysis

[0838] The server receives the resume uploaded by the candidate and converts it into text data using optical character recognition (OCR) technology, then uses natural language processing technology to identify sections of the resume and extract information such as educational background, work history, skills, and motivation for applying.

[0839] Question Set Generation

[0840] Based on the extracted information, the server generates a set of interview questions for the candidate, which are customized according to the company's requirements and the candidate's characteristics.

[0841] Conducting interviews

[0842] The device notifies the candidate of the date, time, and method of the interview. The candidate will be interviewed at the specified date and time using a video conferencing tool (e.g., Zoom). Questions are read aloud by a voice generation AI, and the user (candidate) answers them.

[0843] Analysis of responses

[0844] The server converts the candidate's voice responses into text in real time, which is then analyzed using natural language processing technology to evaluate the specificity and logic of the responses.

[0845] Knowledge test

[0846] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal, and the candidate's answers are sent to the server for evaluation.

[0847] Creating and Providing Feedback

[0848] Based on the results of the interview and knowledge test, the server creates feedback materials, including the candidate's strengths and weaknesses and points to check next time. The created feedback materials are sent to the first interviewer.

[0849] Specific examples of processing

[0850] For example, when a software engineer applicant uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as, "Please tell us about your specific project experience using Python and Django" are generated. On the day of the interview, the voice generation AI on the device reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0851] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0852] The processing flow will be explained below.

[0853] Step 1:

[0854] The server receives resume files uploaded by applicants, which are assumed to be in PDF or Word format.

[0855] Step 2:

[0856] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[0857] Step 3:

[0858] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Specifically, it uses a text analysis algorithm.

[0859] Step 4:

[0860] The server generates a set of questions based on the extracted information, taking into account questions relevant to the position, such as "Tell me about your Python programming experience."

[0861] Step 5:

[0862] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link). Notifications will be sent via email or messaging system.

[0863] Step 6:

[0864] Users are interviewed at the appointed time using a video conferencing tool (such as Zoom), and the device reads out pre-generated questions using speech generation AI.

[0865] Step 7:

[0866] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[0867] Step 8:

[0868] The server converts the user's response voice into text data using speech recognition technology, and the converted text data is analyzed using natural language processing technology.

[0869] Step 9:

[0870] The server evaluates the analyzed answer text and assigns a score based on points such as specificity and logic.

[0871] Step 10:

[0872] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[0873] Step 11:

[0874] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[0875] Step 12:

[0876] The server evaluates the candidate's knowledge test answers and calculates a score.

[0877] Step 13:

[0878] The server combines the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what needs to be checked next time.

[0879] Step 14:

[0880] The server sends the created feedback material to the first interviewer, who uses the material to prepare for the next interview.

[0881] These are the specific steps for efficiently conducting the recruitment process. This system reduces recruitment costs and enables the selection of appropriate personnel.

[0882] Example 1

[0883] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0884] A problem with the traditional recruitment process is that the initial interview stage takes a lot of time and costs a lot of money. Furthermore, evaluations often depend on the interviewer's subjective judgment, making it difficult to objectively evaluate the candidate's aptitude and abilities. For these reasons, there is a need to prevent mismatches and streamline the recruitment process.

[0885] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0886] In this invention, the server includes means for receiving and analyzing resumes, means for converting resumes into text data using optical character recognition technology, means for identifying each section of the resume and extracting information using natural language processing technology, means for generating a set of interview questions based on the extracted information, means for customizing the questions according to the company's requirements and the applicant's characteristics, means for notifying the applicant of the interview date and method, means for conducting the interview using a video conferencing tool and reading the questions aloud using speech generation technology, means for converting the user's voice responses into text, means for analyzing and evaluating the converted text using natural language processing technology, means for creating feedback materials based on the results of the interview and knowledge test, and means for sending the feedback materials to the interviewer, thereby enabling an efficient hiring process and objective evaluation.

[0887] "Means for receiving and analyzing resumes" refers to a series of functions for receiving resumes from applicants and analyzing their contents.

[0888] "Optical character recognition technology" is a technology that extracts character information from documents such as images and PDFs.

[0889] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[0890] "Means for identifying each section of a resume and extracting information" refers to a function that analyzes the contents of a resume to identify each section, such as educational background, work history, and skills, and extracts that information.

[0891] The "means for generating a set of interview questions" is a function that creates questions for candidates based on the analyzed resume information.

[0892] "A means of customizing questions according to the company's requirements and the characteristics of the applicant" refers to a function that appropriately changes questions based on the skills the company is looking for and the background of the applicant.

[0893] "Means of notifying applicants of the interview date and method" is a function that informs applicants of the date and time of the interview and how it will be conducted.

[0894] "Means for conducting interviews using video conferencing tools and reading questions aloud using voice generation technology" refers to a function that conducts interviews using video conferencing software and presents generated questions to candidates as audio.

[0895] The "means for converting the user's voice response into text" is a function for converting the content of the user's verbal response into text data in real time.

[0896] "Means for analyzing and evaluating the converted text using natural language processing technology" refers to a function that analyzes text converted from speech and evaluates the specificity and logic of the content.

[0897] The "means for creating feedback materials" is a function for creating feedback documents for applicants based on the result data of the interview and knowledge test.

[0898] The "means of sending feedback materials to interviewers" is a function for sending the created feedback materials to the first interviewer and other related staff.

[0899] This invention relates to a system that uses artificial intelligence (generative AI models) to conduct interviews in the early stages of the recruitment process, which reduces recruitment costs and prevents mismatches.

[0900] System Overview

[0901] The system is designed to automate the analysis of resumes, generation of question sets, conducting interviews, analyzing responses, conducting knowledge tests, and creating and providing feedback. Each function is described in detail below.

[0902] Resume analysis

[0903] 1. A user accesses the system and uploads a resume.

[0904] 2. The server receives the resume and converts it into text data using optical character recognition (OCR) technology. Specifically, it uses Tesseract OCR.

[0905] 3. The server analyzes the converted text data using natural language processing technology (e.g., spaCy), identifies each section of the resume (educational history, work history, skills, motivation, etc.), and extracts the necessary information.

[0906] Question Set Generation

[0907] 1. The server generates a set of interview questions based on the extracted information, customizing the questions according to the company's requirements and the applicant's characteristics.

[0908] 2. For example, the server uses a generative AI model such as GPT-3 to create questions such as, "Tell me about your specific project experience using Python and Django."

[0909] Conducting interviews

[0910] 1. The device notifies candidates of the date, time, and method of their interview via email, SMS, or a dedicated app.

[0911] 2. Candidates will be interviewed via video conferencing tools (e.g., Zoom) at the appointed date and time.

[0912] 3. During the interview, a voice generation AI (e.g., Google Text-to-Speech) on the device reads out the questions retrieved from the server.

[0913] Analysis of responses

[0914] 1. The candidate answers and the device records the audio.

[0915] 2. The server converts the recorded audio into text in real time using Google Speech-to-Text.

[0916] 3. The converted text is analyzed using natural language processing technology (e.g., spaCy) to evaluate the specificity and logic of the answers.

[0917] Knowledge test

[0918] 1. The server generates knowledge test questions and presents them to the candidate in a predetermined format. The test content includes coding questions and theory questions.

[0919] 2. The candidate answers, and the answer is sent to the server for evaluation.

[0920] Creating and Providing Feedback

[0921] 1. The server creates feedback materials based on the results of the interview and knowledge test. The feedback materials include the candidate's strengths and weaknesses and what needs to be checked next time.

[0922] 2. The prepared feedback materials will be sent to the first interviewer.

[0923] Specific examples

[0924] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as "Tell us about your specific project experience using Python and Django" are generated. On the day of the interview, a voice generation AI reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[0925] Example prompt sentence:

[0926] "Design a system that parses resumes uploaded by candidates, generates interview questions, conducts interviews, parses responses, administers knowledge tests, and generates feedback materials."

[0927] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[0928] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0929] Step 1:

[0930] A user accesses the system and uploads a resume. The input is the user's resume file, and the output is the resume data stored on the server. Specifically, the user creates a profile through the web interface and clicks the resume upload button. The server receives this upload request and saves the resume.

[0931] Step 2:

[0932] The server receives the resume and converts it into text data using optical character recognition (OCR) technology. The input is the uploaded resume file, and the output is the converted text data. Specifically, the server invokes Tesseract OCR to extract text from the resume image.

[0933] Step 3:

[0934] The server uses natural language processing technology to analyze resumes and extract the necessary information. The input is text data, and the output is extracted information such as educational background, work history, skills, and motivation for applying. The server uses spaCy and NLTK to analyze the text and identify keywords and sections.

[0935] Step 4:

[0936] The server generates a set of interview questions based on the extracted information. The input is the extracted information, and the output is the generated set of questions. Specifically, the server retrieves requirements from the company's database and creates questions using a generative AI model (e.g., GPT-3).

[0937] Step 5:

[0938] The server customizes the questions according to the company's requirements and the applicant's characteristics. The input is the generated question set and the company's requirements, and the output is the customized question set. Specifically, the server performs filtering and text fine-tuning on the generated questions.

[0939] Step 6:

[0940] The device notifies the candidate of the interview date, time, and method. The input is information about the interview date, time, and method, and the output is the notified candidate. The device sends this information to the candidate via email, SMS, or a dedicated app.

[0941] Step 7:

[0942] A candidate will be interviewed at a specified date and time using a video conferencing tool. The input is the interview link and date and time information, and the output is joining the video conference. Specifically, the user will log in to the video conferencing tool (e.g., Zoom) using their device.

[0943] Step 8:

[0944] The device uses speech generation AI to read out questions. The input is a customized set of questions, and the output is audio questions. Specifically, the device uses the Google Text-to-Speech API to convert the questions into audio and present them to the candidate as audio played through the speaker.

[0945] Step 9:

[0946] The user (candidate) answers questions, and the device records the audio. The input is the user's voice response, and the output is the recorded audio data. Specifically, the device captures the audio using Zoom's recording function or a dedicated app.

[0947] Step 10:

[0948] The server converts the user's voice response into text. The input is the recorded voice data, and the output is the converted text data. Specifically, the server calls Google Speech-to-Text to convert the voice to text.

[0949] Step 11:

[0950] The server analyzes the converted text and evaluates the specificity and logic of the answer. The input is text data, and the output is the analysis result. Specifically, the server uses spaCy to analyze the text and evaluates it based on specific criteria.

[0951] Step 12:

[0952] The server generates knowledge test questions and presents them to candidates through their terminals. The input is the knowledge test requirements, and the output is a set of knowledge test questions. Specifically, the server uses a question generation algorithm to create coding questions and theory questions and display them on the terminal.

[0953] Step 13:

[0954] The candidate's answers are sent through the terminal and evaluated by the server. The input is the candidate's answer data, and the output is the evaluation result. Specifically, the terminal sends the answer data to the server, which then automatically scores it.

[0955] Step 14:

[0956] The server creates feedback documents based on the results of the interview and knowledge test. The input is the result data of the interview and knowledge test, and the output is the feedback documents. Specifically, the server aggregates the results and generates the feedback documents using a generative AI model.

[0957] Step 15:

[0958] The created feedback material is sent to the first interviewer. The input is the feedback material, and the output is the feedback sent to the first interviewer. In concrete terms, the server sends the created feedback material to the first interviewer via email or an internal system.

[0959] (Application example 1)

[0960] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0961] Maintenance work for factory robots is often complex and requires specialized knowledge, making the judgment of maintenance workers essential. However, because maintenance depends on the skills and experience of the workers, there is a problem of inconsistency in the quality of maintenance. Another issue is the lack of a means to efficiently provide work procedures and provide appropriate feedback in real time, which reduces work efficiency.

[0962] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0963] In this invention, the server includes means for receiving and analyzing a resume, means for identifying each section of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing tool and reading out the questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for sending the feedback materials to the interviewer, means for extracting text information from an image using optical character recognition technology, means for generating a maintenance procedure tailored to a specific situation using a generative artificial intelligence model, and means for evaluating the maintenance work based on the procedure, thereby enabling the efficiency and quality of robot maintenance work in a factory to be improved.

[0964] "Means for receiving and analyzing resumes" refers to a function for electronically receiving resumes submitted by applicants and analyzing their contents.

[0965] "Means for identifying each section of a resume using natural language processing technology" is a function that uses natural language processing technology to classify and identify the contents of a resume into sections such as educational history, work history, and skills.

[0966] The "means for generating a question set based on the analysis results" is a function for generating appropriate interview questions using the analysis results of a resume.

[0967] The "means for conducting an interview using a video conferencing tool and reading out questions" is a function for conducting an interview using a video conferencing tool and reading out the generated questions aloud.

[0968] "Means for converting voice responses into text and analyzing it using natural language processing technology" refers to a function for converting the applicant's voice responses into text data and analyzing that text using natural language processing technology.

[0969] The "means for evaluating the analysis results and creating feedback materials" is a function for evaluating the analysis results of the voice responses and creating materials summarizing feedback to the applicant.

[0970] The "means for sending feedback materials to interviewers" is a function for electronically sending the created feedback materials to interviewers.

[0971] "Means for extracting character information from an image using optical character recognition technology" is a function for reading character information in an image using optical character recognition technology and converting it into text data.

[0972] "Means for generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model" is a function for automatically generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model.

[0973] The "means for evaluating maintenance work based on the procedure" is a function for evaluating work performed based on the generated maintenance procedure.

[0974] This invention relates to a system for improving the efficiency and quality of robot maintenance work in a factory. In one embodiment of the present invention, how a server, a terminal, and a user work together to realize the system will be described.

[0975] Server processing

[0976] Resume analysis

[0977] The server receives the resume uploaded by the user (worker) and converts it into text data using optical character recognition (OCR) technology. The specific software used is OpenCV and pytesseract. After this, each section of the resume (educational history, work history, skills, etc.) is identified using natural language processing technology. The specific software used in this step is Python and a library for natural language processing.

[0978] Question Set Generation

[0979] The server generates a set of interview questions based on the analysis results. This part uses a generative AI model (e.g., GPT-3.5) to create questions that match the characteristics of the candidate and the company's requirements. These questions will also be used to generate maintenance procedures in the future.

[0980] Processing by the terminal

[0981] Conducting interviews

[0982] The device notifies the user of the date, time, and method of the interview. The user then attends the interview at the specified time using a video conferencing tool (e.g., Zoom). The questions are read aloud by a speech generation AI. This part uses Google Cloud Speech-to-Text and Text-to-Speech services for speech recognition and generation technology.

[0983] Analysis of responses

[0984] The server converts the user's voice response into text in real time using Google Cloud Speech-to-Text, and the converted text is analyzed using natural language processing technology to evaluate the specificity and logic of the response.

[0985] Robot Maintenance

[0986] Extracting character information using optical character recognition technology

[0987] The server extracts the robot's state information from the images using optical character recognition techniques, specifically OpenCV and pytesseract.

[0988] Generate maintenance procedures

[0989] The server uses a generative AI model to generate maintenance procedures for specific situations. GPT-3.5 is used to generate these procedures. For example, if the robot's status image displays "Error message 72: Motor overheating," the following prompt sentence is input to the generative AI model:

[0990] The robot status is displayed as "Error message 72: Motor overheated". Please suggest the following maintenance steps:

[0991] Maintenance work evaluation

[0992] The server evaluates the work performed based on the generated maintenance procedure and creates feedback materials based on the evaluation results, which are provided to the workers and used to improve future maintenance work.

[0993] As a result, the efficiency and quality of robot maintenance work within the factory will be improved.

[0994] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0995] Step 1:

[0996] The server receives resumes uploaded by users. The input is the resume uploaded by the user in PDF or image format. The server converts this resume into text data using optical character recognition (OCR) technology. Specifically, it uses OpenCV and pytesseract to recognize characters in the image and extract them as text data. The output is the text data of the resume.

[0997] Step 2:

[0998] The server analyzes the text data extracted by OCR using natural language processing technology. The input is the text data of a resume. The server analyzes this data and identifies each section, such as educational history, work history, and skills. Specifically, it uses Python and natural language processing libraries. The output is the text data of a resume with each section identified.

[0999] Step 3:

[1000] The server generates a set of interview questions based on the analysis results. The input is the text data for each identified section. The server uses a generative AI model (e.g., GPT-3.5) to create a question set that corresponds to the candidate's characteristics and the company's requirements. The output is the generated question set.

[1001] Step 4:

[1002] The terminal notifies the user of the date, time, and method of the interview. The input is the question set and interview schedule data generated from the server. The terminal sends the notification to the user using email or a messaging service. The output is data about the date, time, and method of the notification.

[1003] Step 5:

[1004] The user is interviewed at a specified date and time using a video conferencing tool (e.g., Zoom). The input is a set of questions generated by the server and the user's answers. A voice generation AI on the device reads the questions aloud. The output is the data of the user's voice responses.

[1005] Step 6:

[1006] The server converts the user's voice response into text data in real time. The input is the user's voice response. The server converts the voice to text data using Google Cloud Speech-to-Text technology. The output is the text data response.

[1007] Step 7:

[1008] The server analyzes the text data responses using natural language processing technology. The input is the text data responses. The server analyzes the converted text and evaluates the specificity and logic of the responses. The output is evaluation data of the analysis results.

[1009] Step 8:

[1010] The server creates feedback materials based on the analysis results. The input is the evaluation data. The server generates feedback sentences using a Python script. The output is the feedback materials.

[1011] Step 9:

[1012] The server sends the created feedback materials to the interviewer. The input is the feedback materials. The server sends the materials via email. The output is a notification that the materials have been sent.

[1013] Step 10:

[1014] The server uses optical character recognition technology to extract robot status information from images. The input is image data showing the robot's status. The server uses OpenCV and pytesseract to convert the character information in the image into text data. The output is text data of the robot's status.

[1015] Step 11:

[1016] The server uses a generative artificial intelligence model to generate maintenance procedures tailored to specific situations. The input is text data of the robot's status. The server uses a generative AI model (e.g., GPT-3.5) to generate appropriate maintenance procedures. For example, the prompt sentence "The robot's status is displayed as 'Error message 72: Motor overheating'. Please suggest the next maintenance procedure." is input to the model. The output is the generated maintenance procedure.

[1017] Step 12:

[1018] The server evaluates the maintenance work based on the procedure. The input is the generated maintenance procedure and the maintenance work result data. The server evaluates the work performed according to the procedure and creates the results as feedback material. The output is the evaluation result of the maintenance work and the feedback material.

[1019] Step 13:

[1020] The server sends the created feedback material to the worker. The input is the feedback material. The server sends the material via email. The output is a notification that the material has been sent.

[1021] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1022] System Overview

[1023] This invention is a system that uses artificial intelligence (generative AI models) in the early stages of the hiring process. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate hiring process.

[1024] Resume analysis

[1025] The server receives resumes uploaded by candidates, converts the resume files (PDF or Word files) into text data using optical character recognition (OCR), and then uses natural language processing (NLP) to extract information such as educational background, work history, skills, and motivation for applying from the resume text.

[1026] Question Set Generation

[1027] The server generates a set of questions based on the extracted information, customized to include content relevant to the position being applied for. For example, a software engineer candidate might be asked questions like, "Tell me about your Python programming experience."

[1028] Conducting interviews

[1029] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). The candidate takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions. The user (candidate) answers them.

[1030] Incorporating an emotion engine

[1031] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is then sent to the server.

[1032] Analysis of response and sentiment data

[1033] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. The response is evaluated based on factors such as specificity and logic. Emotion data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed.

[1034] Knowledge test

[1035] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[1036] Creating and Providing Feedback

[1037] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[1038] Specific examples of processing

[1039] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. During the interview, they are asked questions such as, "Tell us about your specific project experience using Python and Django," and the emotion engine analyzes the user's facial expressions and tone of voice while answering. The server analyzes the voice responses and emotion data and makes an evaluation such as, "The answers were highly specific and emotionally confident." This evaluation is reflected in the feedback materials and sent to the first interviewer.

[1040] By incorporating emotion recognition technology, this system enables objective evaluation that is not influenced by the interviewer's subjectivity, contributing to the selection of more appropriate personnel. It also improves convenience for applicants by providing flexible interview opportunities.

[1041] The processing flow will be explained below.

[1042] Step 1:

[1043] The server receives resume files uploaded by applicants, which are expected to be in PDF or Word format.

[1044] Step 2:

[1045] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[1046] Step 3:

[1047] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data.

[1048] Step 4:

[1049] The server generates a set of questions based on the extracted information, including questions such as "Tell me about your Python programming experience."

[1050] Step 5:

[1051] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link) via email or messaging system.

[1052] Step 6:

[1053] The user will be interviewed at the appointed date and time using a video conferencing tool (Zoom), with the device reading out pre-generated questions using a speech-generating AI.

[1054] Step 7:

[1055] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[1056] Step 8:

[1057] During the interview, the device activates an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time, which is then sent to the server.

[1058] Step 9:

[1059] The server uses speech recognition technology to convert the user's voice response into text data. The converted text is then analyzed using natural language processing technology to evaluate points such as the specificity and logic of the response. At the same time, emotional data obtained from an emotion recognition engine is also analyzed.

[1060] Step 10:

[1061] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[1062] Step 11:

[1063] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[1064] Step 12:

[1065] The server evaluates the candidate's knowledge test answers and calculates a score.

[1066] Step 13:

[1067] The server aggregates all responses and emotional data to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. The emotional data indicates the candidate's emotional changes during the interview and their overall emotional state.

[1068] Step 14:

[1069] The server transmits the created feedback material to the first interviewer, who can use this material to efficiently prepare for the next interview.

[1070] Example 2

[1071] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1072] In the traditional hiring process, it is easily influenced by the interviewer's subjectivity, making it difficult to make an objective evaluation. Furthermore, there is no way to evaluate the candidate's emotional changes and facial expressions in real time, making it difficult to make an accurate and fair evaluation. Furthermore, the typical interview process is time-consuming, labor-intensive, and inefficient.

[1073] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1074] In this invention, the server includes means for receiving and analyzing resumes, means for converting resume files into text data using optical character recognition technology, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for inputting prompts and generating appropriate questions using a generative AI model, means for conducting interviews using a video conferencing tool and reading out questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for analyzing facial expressions and tone of voice during the interview using an emotion recognition engine and extracting emotion data, means for integrating the analysis results and the emotion data and evaluating them, means for creating feedback materials, and means for sending the feedback materials to interviewers. This enables objective and accurate evaluation, including changes in the emotions of candidates, and improves the efficiency and fairness of the hiring process.

[1075] "Means for receiving and analyzing resumes" refers to the device or software used to receive resume files uploaded by applicants and analyze their contents.

[1076] "Optical character recognition technology" is a technology that extracts text information from image data and is used to convert digital resume files into text data.

[1077] "Natural language processing technology" is a technology that uses computers to analyze and understand text written in natural language, and is used to extract necessary information from text data such as resumes.

[1078] A "means for generating a question set" is a device or software for generating appropriate interview questions based on the analyzed resume data.

[1079] A "generative AI model" is a trained artificial intelligence model that has the ability to generate natural language when given a prompt sentence.

[1080] A "prompt" is an instruction given to an AI model to perform a specific task.

[1081] A "video conferencing tool" is software for sharing audio and video in real time over the Internet.

[1082] A "means for converting voice responses to text" is a device or software that uses voice recognition technology to convert voice responses during an interview into text form.

[1083] An "emotion recognition engine" is a device or software that analyzes a person's emotional state from audio and video data and extracts emotional data in real time.

[1084] The "means for creating feedback materials" refers to a device or software for generating feedback materials summarizing the evaluation of candidates based on the results of the interview and knowledge test.

[1085] The "means for sending feedback materials to interviewers" refers to a device or software for automatically sending the created feedback materials to interviewers.

[1086] This invention is a system that combines artificial intelligence (generative AI models) and an emotion recognition engine to perform more accurate evaluations in the early stages of the hiring process. This system automates the processes of resume analysis, question generation, interviews, response analysis, knowledge testing, and feedback creation.

[1087] Hardware and software used

[1088] The system uses the following hardware and software:

[1089] 1. Server

[1090] A server with the storage and processing power to receive resumes

[1091] Optical Character Recognition (OCR) engine (e.g., Tesseract)

[1092] Natural Language Processing (NLP) engines (e.g., SpaCy, BERT)

[1093] Generative AI models (e.g., GPT-3)

[1094] Voice recognition technology (e.g., Google Speech-to-Text)

[1095] Report generation tools (e.g. Tableau)

[1096] 2. Terminal

[1097] Devices for real-time communication with users (PCs, smartphones, tablets, etc.)

[1098] Video conferencing tools (e.g., Zoom, Microsoft Teams)

[1099] Voice generation AI (e.g. Amazon Polly)

[1100] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[1101] 3. Users

[1102] Applicants and interviewers

[1103] Specific processing of the program

[1104] Resume upload and analysis

[1105] Users upload their resumes to the server, which converts the received resume files (PDF or Word files) into text data using an optical character recognition (OCR) engine. It then uses natural language processing (NLP) to extract information such as educational background, work history, and skills from the resume text data.

[1106] Question Set Generation

[1107] The server generates a set of relevant questions for the candidate based on the analyzed resume data. It uses a generative AI model (GPT-3) to generate appropriate questions. For example, the server inputs a prompt such as, "Generate interview questions based on the candidate's programming skills."

[1108] Preparing for and conducting interviews

[1109] The device notifies the candidate of the date, time, and method of the interview, sends a link to a video conferencing tool (such as Zoom), and the interview takes place at the specified date and time. During the interview, a speech-generation AI (Amazon Polly) is used to read out pre-generated questions.

[1110] Analysis by emotion engine

[1111] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice, extracting emotional data in real time and sending it to the server.

[1112] Analysis of response and sentiment data

[1113] The server converts the applicant's voice responses into text data using speech recognition technology and analyzes them using natural language processing technology. The server evaluates the specificity and logic of the responses. It also analyzes emotional data obtained from an emotion recognition engine to evaluate emotional changes.

[1114] Knowledge test

[1115] The server creates knowledge test questions using a generative AI model and presents them to the candidate via their device. For example, the candidate might enter a prompt such as, "Please create a question about data structures and algorithms." The candidate's answers are then sent to the server for evaluation.

[1116] Creating and Providing Feedback

[1117] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. This feedback includes emotional data and is evaluated in detail. The created feedback materials are sent to the first interviewer.

[1118] Examples of prompt statements

[1119] 1. "Please parse this resume and extract Python and Django project experience."

[1120] 2. "Generate questions to assess knowledge of data structures and algorithms."

[1121] 3. "Evaluate the specificity of interview responses and create feedback materials that analyze emotional data."

[1122] By integrating emotion recognition and AI technology, the system eliminates subjective interviewer evaluations, significantly improving the accuracy and efficiency of the hiring process, while providing candidates with a flexible interview experience for increased convenience.

[1123] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1124] Step 1:

[1125] Resume upload and analysis

[1126] The user uploads their resume to the system. The server stores the received resume file (PDF or Word file) in storage. Optical character recognition (OCR) technology is used to convert the file into text data. The input is a PDF or Word file, and the output is text data. The server launches an OCR engine (e.g., Tesseract) to analyze the file and extract the text. Next, natural language processing (NLP) technology is used to extract information such as education, work history, and skills from the resume text data. An NLP engine (e.g., SpaCy) is used to identify the information for each section.

[1127] Step 2:

[1128] Question Set Generation

[1129] The server generates a set of questions relevant to the candidate based on the analyzed information. It uses a generative AI model (e.g., GPT-3) and inputs a prompt to generate questions. The input is the analyzed resume data and the prompt, and the output is a set of interview questions. For example, the prompt "Generate interview questions based on the candidate's programming skills" is input to the generative AI model. The generated questions are saved for interview preparation.

[1130] Step 3:

[1131] Preparing for and conducting interviews

[1132] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). Notification is sent via email or SMS. The input is the interview date, time, and method, and the output is the notification to the candidate. The user (candidate) takes the interview using a video conferencing tool at the notified date and time. During the interview, the device uses a speech generation AI (e.g., Amazon Polly) to read out pre-generated questions. The input is the generated question set, and the output is the audio reading.

[1133] Step 4:

[1134] Analysis by emotion engine

[1135] During the interview, the device runs an emotion recognition engine to analyze the user's (candidate's) facial expressions and tone of voice. Emotional data is extracted in real time and sent to the server. The input is video and audio data, and the output is emotion data. Specifically, a webcam and microphone are used to capture face and voice, and the emotion recognition engine (e.g., Microsoft Azure Emotion API) analyzes this data in real time.

[1136] Step 5:

[1137] Analysis of response and sentiment data

[1138] The server converts the candidate's voice response into text data using speech recognition technology (e.g., Google Speech-to-Text). It then analyzes the response using natural language processing technology. The input is the voice response and emotional data, and the output is the analyzed text and emotional evaluation. The specificity and logic of the response are evaluated, and data obtained from the emotion recognition engine is simultaneously analyzed to evaluate emotional changes. The text data is evaluated using an NLP engine (e.g., BERT).

[1139] Step 6:

[1140] Knowledge test

[1141] The server uses a generative AI model to create knowledge test questions and presents them to the candidate via their terminal. For example, a prompt such as "Please create questions about data structures and algorithms" is input to the generative AI model. The input is the prompt to the generative AI model, and the output is the knowledge test questions. The user answers the test, and the answers are sent back to the server. The input is the candidate's test answers, and the output is the evaluated answers.

[1142] Step 7:

[1143] Creating and Providing Feedback

[1144] The server integrates the results of the interview and knowledge test and creates feedback materials that include the candidate's strengths and weaknesses and what to check next time. The feedback also includes emotional data. The inputs are the interview results, knowledge test results, and emotional data, and the output is the feedback materials. The created feedback materials are sent to the first interviewer. Specifically, a report generation tool (e.g., Tableau) is used to provide feedback in a visually easy-to-understand format.

[1145] (Application example 2)

[1146] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1147] Traditional hiring processes and customer service in brick-and-mortar stores are prone to errors, time-consuming manual operations, and subjective evaluation biases by interviewers and store clerks. Furthermore, traditional systems are unable to recognize users' emotions in real time and respond accordingly. This creates a need for improved hiring efficiency and customer satisfaction.

[1148] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1149] In this invention, the server includes means for receiving and analyzing a resume, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing engine and reading out questions, means for converting voice responses to text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for transmitting the feedback materials to the interviewer, an emotion recognition engine for extracting emotion data, means for analyzing the emotion data to evaluate the user's emotional state, means for answering the user's questions in real time, means for using natural language processing technology to understand the content of the user's questions, and means for analyzing and displaying data input to smart glasses or a head-mounted display. This enables efficient and objective evaluation in the recruitment process and customer service in physical stores, ultimately improving user satisfaction.

[1150] A "resume" is a document submitted by an applicant that includes information such as a self-introduction, educational background, work history, skills, and motivation for applying.

[1151] "Analysis" is the process of breaking down data or information to understand its content for a specific purpose.

[1152] "Natural language processing technology" is technology that enables computers to understand, interpret, and generate human language.

[1153] A "section" is a part of a document or data that is divided based on a specific theme or content.

[1154] A "question set" is a collection containing a series of questions created for a specific purpose.

[1155] A "video conferencing engine" is software or a system for visual and audio communication over the Internet.

[1156] A "voice response" is a response or reply given aloud.

[1157] "Convert to text" is the process of converting non-text data, such as audio or images, into character data.

[1158] "Feedback materials" are documents that compile information including the results of the evaluation and areas for improvement next time.

[1159] An "interviewer" is a person responsible for conducting an interview with an applicant and evaluating the results.

[1160] An "emotion recognition engine" is a system that detects and recognizes emotions from facial expressions, tone of voice, body movements, etc.

[1161] "Emotional state" refers to a person's emotional or psychological state at a given time.

[1162] "Smart glasses" are glasses with additional functions, such as displaying information and providing augmented reality functions.

[1163] A "head-mounted display" is a display device worn on the head to display images in front of the eyes.

[1164] System configuration

[1165] This invention is a system that combines artificial intelligence and emotion recognition technology for the early stages of the hiring process and customer service in brick-and-mortar stores. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate process. It can also be applied to real-time customer service in brick-and-mortar stores.

[1166] Resume analysis

[1167] The server receives resumes uploaded by applicants. It uses optical character recognition (OCR) technology to convert the resume file (PDF or Word file) into text data. It then uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Examples of software used here include Tesseract OCR and spaCy.

[1168] Question Set Generation

[1169] The server generates a set of questions based on the extracted information, customized to include content relevant to the position. For example, a generative AI model could be used to generate questions like, "Tell me about your Python programming experience."

[1170] Conducting interviews

[1171] The device notifies the applicant of the date, time, and method of the interview (for example, a Zoom link using a video conferencing engine). The applicant takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions, and the user (applicant) answers them.

[1172] Incorporating an emotion engine

[1173] During the interview, the device runs an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is sent to a server for analysis. For example, OpenCV or EmotionRecognizer is used.

[1174] Analysis of response and sentiment data

[1175] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. Points such as the specificity and logic of the response are also evaluated. Emotional data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed. As a result, if the user's response is highly specific and emotionally confident, it can be evaluated as an "excellent answer."

[1176] Knowledge test

[1177] The server generates knowledge test questions in a predetermined format and presents them to the applicant via the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[1178] Creating and Providing Feedback

[1179] The server integrates the results of the interview and knowledge test to create feedback materials that include the applicant's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[1180] Application in physical stores

[1181] This system can also be applied to customer service in brick-and-mortar stores. For example, when a customer asks a question, a response can be made in real time. An emotion recognition engine analyzes the video input and evaluates customer satisfaction. This makes it possible to respond to customer needs quickly, which is expected to improve customer satisfaction.

[1182] Prompt Sentence Examples

[1183] "When a customer asks, 'Do you have this shirt in size large?', please provide an appropriate response based on your product inventory data."

[1184] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1185] Processing Steps of the Detailed Description

[1186] Step 1:

[1187] The server receives resumes uploaded by applicants. The submitted resumes are in PDF or Word file format and are converted into text data using optical character recognition (OCR) technology. Specifically, Tesseract OCR is used to analyze the characters in the image and generate text data.

[1188] Step 2:

[1189] The server analyzes the text data using natural language processing (NLP) technology to identify each section, such as educational background, work history, skills, motivation, etc. At this stage, the text is grammatically analyzed using, for example, spaCy, to extract the necessary information.

[1190] Step 3:

[1191] The server generates a set of questions based on the extracted information. Using a generative AI model, it generates questions relevant to the position being applied for. For example, specific questions such as "Tell me about your Python programming experience" are generated based on the prompt. This set of questions is output as text data.

[1192] Step 4:

[1193] The device notifies the applicant of the interview date and method (such as a Zoom link using a video conferencing engine). The notification is sent to the applicant in the form of a message or email. The user (applicant) receives the notification.

[1194] Step 5:

[1195] The user (applicant) is interviewed using a video conferencing tool. During the interview, the device uses a speech generation AI to read out pre-generated questions based on the question set generated in Step 3.

[1196] Step 6:

[1197] The user (applicant) answers questions by voice. The device collects the voice responses in real time and sends the voice data to the server.

[1198] Step 7:

[1199] The server converts the user's voice response into text data using voice recognition technology, such as the Google Speech-to-Text API.

[1200] Step 8:

[1201] The server then analyzes the converted text data using natural language processing technology to evaluate the specificity and logic of the answers. At this stage, text analysis is again performed using tools such as spaCy.

[1202] Step 9:

[1203] During the interview, the device runs an emotion recognition engine, analyzes the user's facial expressions and tone of voice, and extracts emotional data in real time. This emotional data is input as image and audio data and analyzed using OpenCV and EmotionRecognizer.

[1204] Step 10:

[1205] The server analyzes the emotional data and evaluates the user's emotional state. Specifically, it evaluates the user's emotional changes over time and quantifies their confidence and nervousness during the interview.

[1206] Step 11:

[1207] The server combines the analysis results of the interview voice responses with the emotional data to create feedback materials, which include the user's strengths and weaknesses and points to check next time. The feedback materials are output in text format.

[1208] Step 12:

[1209] The server sends the created feedback materials to the interviewer, who is then provided with the materials via email or a dedicated portal.

[1210] Step 13:

[1211] The server generates questions for the knowledge test in a predetermined format and presents them to the applicant via the terminal. The questions for the knowledge test are entered in text format, and the user answers them.

[1212] Step 14:

[1213] The user answers the knowledge test questions and sends the answer data to the server, which evaluates the knowledge test answers and reflects the results in the feedback material.

[1214] Step 15:

[1215] In a physical store, the device uses natural language processing technology and generative AI models to answer user questions in real time. For example, if a user asks, "Do you have this shirt in size L?", the device will provide an appropriate answer based on inventory data.

[1216] Step 16:

[1217] The device uses smart glasses or a head-mounted display to analyze the input data and provide visual information to the user. The user's visual data is analyzed in real time, and appropriate product information and guidance are displayed.

[1218] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1219] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1220] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1221] [Fourth embodiment]

[1222] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1223] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1224] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1225] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1226] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1227] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1228] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1229] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1230] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1231] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1232] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1233] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1234] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1235] System Overview

[1236] This invention is a system that uses artificial intelligence (generative AI models) to conduct interviews at the early stage of the recruitment process, reducing recruitment costs and preventing mismatches.

[1237] Resume analysis

[1238] The server receives the resume uploaded by the candidate and converts it into text data using optical character recognition (OCR) technology, then uses natural language processing technology to identify sections of the resume and extract information such as educational background, work history, skills, and motivation for applying.

[1239] Question Set Generation

[1240] Based on the extracted information, the server generates a set of interview questions for the candidate, which are customized according to the company's requirements and the candidate's characteristics.

[1241] Conducting interviews

[1242] The device notifies the candidate of the date, time, and method of the interview. The candidate will be interviewed at the specified date and time using a video conferencing tool (e.g., Zoom). Questions are read aloud by a voice generation AI, and the user (candidate) answers them.

[1243] Analysis of responses

[1244] The server converts the candidate's voice responses into text in real time, which is then analyzed using natural language processing technology to evaluate the specificity and logic of the responses.

[1245] Knowledge test

[1246] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal, and the candidate's answers are sent to the server for evaluation.

[1247] Creating and Providing Feedback

[1248] Based on the results of the interview and knowledge test, the server creates feedback materials, including the candidate's strengths and weaknesses and points to check next time. The created feedback materials are sent to the first interviewer.

[1249] Specific examples of processing

[1250] For example, when a software engineer applicant uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as, "Please tell us about your specific project experience using Python and Django" are generated. On the day of the interview, the voice generation AI on the device reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[1251] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[1252] The processing flow will be explained below.

[1253] Step 1:

[1254] The server receives resume files uploaded by applicants, which are assumed to be in PDF or Word format.

[1255] Step 2:

[1256] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[1257] Step 3:

[1258] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Specifically, it uses a text analysis algorithm.

[1259] Step 4:

[1260] The server generates a set of questions based on the extracted information, taking into account questions relevant to the position, such as "Tell me about your Python programming experience."

[1261] Step 5:

[1262] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link). Notifications will be sent via email or messaging system.

[1263] Step 6:

[1264] Users are interviewed at the appointed time using a video conferencing tool (such as Zoom), and the device reads out pre-generated questions using speech generation AI.

[1265] Step 7:

[1266] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[1267] Step 8:

[1268] The server converts the user's response voice into text data using speech recognition technology, and the converted text data is analyzed using natural language processing technology.

[1269] Step 9:

[1270] The server evaluates the analyzed answer text and assigns a score based on points such as specificity and logic.

[1271] Step 10:

[1272] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[1273] Step 11:

[1274] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[1275] Step 12:

[1276] The server evaluates the candidate's knowledge test answers and calculates a score.

[1277] Step 13:

[1278] The server combines the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what needs to be checked next time.

[1279] Step 14:

[1280] The server sends the created feedback material to the first interviewer, who uses the material to prepare for the next interview.

[1281] These are the specific steps for efficiently conducting the recruitment process. This system reduces recruitment costs and enables the selection of appropriate personnel.

[1282] Example 1

[1283] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1284] A problem with the traditional recruitment process is that the initial interview stage takes a lot of time and costs a lot of money. Furthermore, evaluations often depend on the interviewer's subjective judgment, making it difficult to objectively evaluate the candidate's aptitude and abilities. For these reasons, there is a need to prevent mismatches and streamline the recruitment process.

[1285] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1286] In this invention, the server includes means for receiving and analyzing resumes, means for converting resumes into text data using optical character recognition technology, means for identifying each section of the resume and extracting information using natural language processing technology, means for generating a set of interview questions based on the extracted information, means for customizing the questions according to the company's requirements and the applicant's characteristics, means for notifying the applicant of the interview date and method, means for conducting the interview using a video conferencing tool and reading the questions aloud using speech generation technology, means for converting the user's voice responses into text, means for analyzing and evaluating the converted text using natural language processing technology, means for creating feedback materials based on the results of the interview and knowledge test, and means for sending the feedback materials to the interviewer, thereby enabling an efficient hiring process and objective evaluation.

[1287] "Means for receiving and analyzing resumes" refers to a series of functions for receiving resumes from applicants and analyzing their contents.

[1288] "Optical character recognition technology" is a technology that extracts character information from documents such as images and PDFs.

[1289] "Natural language processing technology" is a technology that allows computers to understand and analyze human language.

[1290] "Means for identifying each section of a resume and extracting information" refers to a function that analyzes the contents of a resume to identify each section, such as educational background, work history, and skills, and extracts that information.

[1291] The "means for generating a set of interview questions" is a function that creates questions for candidates based on the analyzed resume information.

[1292] "A means of customizing questions according to the company's requirements and the characteristics of the applicant" refers to a function that appropriately changes questions based on the skills the company is looking for and the background of the applicant.

[1293] "Means of notifying applicants of the interview date and method" is a function that informs applicants of the date and time of the interview and how it will be conducted.

[1294] "Means for conducting interviews using video conferencing tools and reading questions aloud using voice generation technology" refers to a function that conducts interviews using video conferencing software and presents generated questions to candidates as audio.

[1295] The "means for converting the user's voice response into text" is a function for converting the content of the user's verbal response into text data in real time.

[1296] "Means for analyzing and evaluating the converted text using natural language processing technology" refers to a function that analyzes text converted from speech and evaluates the specificity and logic of the content.

[1297] The "means for creating feedback materials" is a function for creating feedback documents for applicants based on the result data of the interview and knowledge test.

[1298] The "means of sending feedback materials to interviewers" is a function for sending the created feedback materials to the first interviewer and other related staff.

[1299] This invention relates to a system that uses artificial intelligence (generative AI models) to conduct interviews in the early stages of the recruitment process, which reduces recruitment costs and prevents mismatches.

[1300] System Overview

[1301] The system is designed to automate the analysis of resumes, generation of question sets, conducting interviews, analyzing responses, conducting knowledge tests, and creating and providing feedback. Each function is described in detail below.

[1302] Resume analysis

[1303] 1. A user accesses the system and uploads a resume.

[1304] 2. The server receives the resume and converts it into text data using optical character recognition (OCR) technology. Specifically, it uses Tesseract OCR.

[1305] 3. The server analyzes the converted text data using natural language processing technology (e.g., spaCy), identifies each section of the resume (educational history, work history, skills, motivation, etc.), and extracts the necessary information.

[1306] Question Set Generation

[1307] 1. The server generates a set of interview questions based on the extracted information, customizing the questions according to the company's requirements and the applicant's characteristics.

[1308] 2. For example, the server uses a generative AI model such as GPT-3 to create questions such as, "Tell me about your specific project experience using Python and Django."

[1309] Conducting interviews

[1310] 1. The device notifies candidates of the date, time, and method of their interview via email, SMS, or a dedicated app.

[1311] 2. Candidates will be interviewed via video conferencing tools (e.g., Zoom) at the appointed date and time.

[1312] 3. During the interview, a voice generation AI (e.g., Google Text-to-Speech) on the device reads out the questions retrieved from the server.

[1313] Analysis of responses

[1314] 1. The candidate answers and the device records the audio.

[1315] 2. The server converts the recorded audio into text in real time using Google Speech-to-Text.

[1316] 3. The converted text is analyzed using natural language processing technology (e.g., spaCy) to evaluate the specificity and logic of the answers.

[1317] Knowledge test

[1318] 1. The server generates knowledge test questions and presents them to the candidate in a predetermined format. The test content includes coding questions and theory questions.

[1319] 2. The candidate answers, and the answer is sent to the server for evaluation.

[1320] Creating and Providing Feedback

[1321] 1. The server creates feedback materials based on the results of the interview and knowledge test. The feedback materials include the candidate's strengths and weaknesses and what needs to be checked next time.

[1322] 2. The prepared feedback materials will be sent to the first interviewer.

[1323] Specific examples

[1324] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. Then, questions such as "Tell us about your specific project experience using Python and Django" are generated. On the day of the interview, a voice generation AI reads out these questions, and the user answers. The server analyzes and evaluates the answers, then creates feedback materials and provides them to the first interviewer.

[1325] Example prompt sentence:

[1326] "Design a system that parses resumes uploaded by candidates, generates interview questions, conducts interviews, parses responses, administers knowledge tests, and generates feedback materials."

[1327] This system will streamline the hiring process, enable objective evaluation that is not influenced by the interviewer's subjectivity, and improve convenience by allowing candidates to take interviews at their own convenience.

[1328] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1329] Step 1:

[1330] A user accesses the system and uploads a resume. The input is the user's resume file, and the output is the resume data stored on the server. Specifically, the user creates a profile through the web interface and clicks the resume upload button. The server receives this upload request and saves the resume.

[1331] Step 2:

[1332] The server receives the resume and converts it into text data using optical character recognition (OCR) technology. The input is the uploaded resume file, and the output is the converted text data. Specifically, the server invokes Tesseract OCR to extract text from the resume image.

[1333] Step 3:

[1334] The server uses natural language processing technology to analyze resumes and extract the necessary information. The input is text data, and the output is extracted information such as educational background, work history, skills, and motivation for applying. The server uses spaCy and NLTK to analyze the text and identify keywords and sections.

[1335] Step 4:

[1336] The server generates a set of interview questions based on the extracted information. The input is the extracted information, and the output is the generated set of questions. Specifically, the server retrieves requirements from the company's database and creates questions using a generative AI model (e.g., GPT-3).

[1337] Step 5:

[1338] The server customizes the questions according to the company's requirements and the applicant's characteristics. The input is the generated question set and the company's requirements, and the output is the customized question set. Specifically, the server performs filtering and text fine-tuning on the generated questions.

[1339] Step 6:

[1340] The device notifies the candidate of the interview date, time, and method. The input is information about the interview date, time, and method, and the output is the notified candidate. The device sends this information to the candidate via email, SMS, or a dedicated app.

[1341] Step 7:

[1342] A candidate will be interviewed at a specified date and time using a video conferencing tool. The input is the interview link and date and time information, and the output is joining the video conference. Specifically, the user will log in to the video conferencing tool (e.g., Zoom) using their device.

[1343] Step 8:

[1344] The device uses speech generation AI to read out questions. The input is a customized set of questions, and the output is audio questions. Specifically, the device uses the Google Text-to-Speech API to convert the questions into audio and present them to the candidate as audio played through the speaker.

[1345] Step 9:

[1346] The user (candidate) answers questions, and the device records the audio. The input is the user's voice response, and the output is the recorded audio data. Specifically, the device captures the audio using Zoom's recording function or a dedicated app.

[1347] Step 10:

[1348] The server converts the user's voice response into text. The input is the recorded voice data, and the output is the converted text data. Specifically, the server calls Google Speech-to-Text to convert the voice to text.

[1349] Step 11:

[1350] The server analyzes the converted text and evaluates the specificity and logic of the answer. The input is text data, and the output is the analysis result. Specifically, the server uses spaCy to analyze the text and evaluates it based on specific criteria.

[1351] Step 12:

[1352] The server generates knowledge test questions and presents them to candidates through their terminals. The input is the knowledge test requirements, and the output is a set of knowledge test questions. Specifically, the server uses a question generation algorithm to create coding questions and theory questions and display them on the terminal.

[1353] Step 13:

[1354] The candidate's answers are sent through the terminal and evaluated by the server. The input is the candidate's answer data, and the output is the evaluation result. Specifically, the terminal sends the answer data to the server, which then automatically scores it.

[1355] Step 14:

[1356] The server creates feedback documents based on the results of the interview and knowledge test. The input is the result data of the interview and knowledge test, and the output is the feedback documents. Specifically, the server aggregates the results and generates the feedback documents using a generative AI model.

[1357] Step 15:

[1358] The created feedback material is sent to the first interviewer. The input is the feedback material, and the output is the feedback sent to the first interviewer. In concrete terms, the server sends the created feedback material to the first interviewer via email or an internal system.

[1359] (Application example 1)

[1360] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1361] Maintenance work for factory robots is often complex and requires specialized knowledge, making the judgment of maintenance workers essential. However, because maintenance depends on the skills and experience of the workers, there is a problem of inconsistency in the quality of maintenance. Another issue is the lack of a means to efficiently provide work procedures and provide appropriate feedback in real time, which reduces work efficiency.

[1362] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1363] In this invention, the server includes means for receiving and analyzing a resume, means for identifying each section of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing tool and reading out the questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for sending the feedback materials to the interviewer, means for extracting text information from an image using optical character recognition technology, means for generating a maintenance procedure tailored to a specific situation using a generative artificial intelligence model, and means for evaluating the maintenance work based on the procedure, thereby enabling the efficiency and quality of robot maintenance work in a factory to be improved.

[1364] "Means for receiving and analyzing resumes" refers to a function for electronically receiving resumes submitted by applicants and analyzing their contents.

[1365] "Means for identifying each section of a resume using natural language processing technology" is a function that uses natural language processing technology to classify and identify the contents of a resume into sections such as educational history, work history, and skills.

[1366] The "means for generating a question set based on the analysis results" is a function for generating appropriate interview questions using the analysis results of a resume.

[1367] The "means for conducting an interview using a video conferencing tool and reading out questions" is a function for conducting an interview using a video conferencing tool and reading out the generated questions aloud.

[1368] "Means for converting voice responses into text and analyzing it using natural language processing technology" refers to a function for converting the applicant's voice responses into text data and analyzing that text using natural language processing technology.

[1369] The "means for evaluating the analysis results and creating feedback materials" is a function for evaluating the analysis results of the voice responses and creating materials summarizing feedback to the applicant.

[1370] The "means for sending feedback materials to interviewers" is a function for electronically sending the created feedback materials to interviewers.

[1371] "Means for extracting character information from an image using optical character recognition technology" is a function for reading character information in an image using optical character recognition technology and converting it into text data.

[1372] "Means for generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model" is a function for automatically generating a maintenance procedure suited to a specific situation using a generative artificial intelligence model.

[1373] The "means for evaluating maintenance work based on the procedure" is a function for evaluating work performed based on the generated maintenance procedure.

[1374] This invention relates to a system for improving the efficiency and quality of robot maintenance work in a factory. In one embodiment of the present invention, how a server, a terminal, and a user work together to realize the system will be described.

[1375] Server processing

[1376] Resume analysis

[1377] The server receives the resume uploaded by the user (worker) and converts it into text data using optical character recognition (OCR) technology. The specific software used is OpenCV and pytesseract. After this, each section of the resume (educational history, work history, skills, etc.) is identified using natural language processing technology. The specific software used in this step is Python and a library for natural language processing.

[1378] Question Set Generation

[1379] The server generates a set of interview questions based on the analysis results. This part uses a generative AI model (e.g., GPT-3.5) to create questions that match the characteristics of the candidate and the company's requirements. These questions will also be used to generate maintenance procedures in the future.

[1380] Processing by the terminal

[1381] Conducting interviews

[1382] The device notifies the user of the date, time, and method of the interview. The user then attends the interview at the specified time using a video conferencing tool (e.g., Zoom). The questions are read aloud by a speech generation AI. This part uses Google Cloud Speech-to-Text and Text-to-Speech services for speech recognition and generation technology.

[1383] Analysis of responses

[1384] The server converts the user's voice response into text in real time using Google Cloud Speech-to-Text, and the converted text is analyzed using natural language processing technology to evaluate the specificity and logic of the response.

[1385] Robot Maintenance

[1386] Extracting character information using optical character recognition technology

[1387] The server extracts the robot's state information from the images using optical character recognition techniques, specifically OpenCV and pytesseract.

[1388] Generate maintenance procedures

[1389] The server uses a generative AI model to generate maintenance procedures for specific situations. GPT-3.5 is used to generate these procedures. For example, if the robot's status image displays "Error message 72: Motor overheating," the following prompt sentence is input to the generative AI model:

[1390] The robot status is displayed as "Error message 72: Motor overheated". Please suggest the following maintenance steps:

[1391] Maintenance work evaluation

[1392] The server evaluates the work performed based on the generated maintenance procedure and creates feedback materials based on the evaluation results, which are provided to the workers and used to improve future maintenance work.

[1393] As a result, the efficiency and quality of robot maintenance work within the factory will be improved.

[1394] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1395] Step 1:

[1396] The server receives resumes uploaded by users. The input is the resume uploaded by the user in PDF or image format. The server converts this resume into text data using optical character recognition (OCR) technology. Specifically, it uses OpenCV and pytesseract to recognize characters in the image and extract them as text data. The output is the text data of the resume.

[1397] Step 2:

[1398] The server analyzes the text data extracted by OCR using natural language processing technology. The input is the text data of a resume. The server analyzes this data and identifies each section, such as educational history, work history, and skills. Specifically, it uses Python and natural language processing libraries. The output is the text data of a resume with each section identified.

[1399] Step 3:

[1400] The server generates a set of interview questions based on the analysis results. The input is the text data for each identified section. The server uses a generative AI model (e.g., GPT-3.5) to create a question set that corresponds to the candidate's characteristics and the company's requirements. The output is the generated question set.

[1401] Step 4:

[1402] The terminal notifies the user of the date, time, and method of the interview. The input is the question set and interview schedule data generated from the server. The terminal sends the notification to the user using email or a messaging service. The output is data about the date, time, and method of the notification.

[1403] Step 5:

[1404] The user is interviewed at a specified date and time using a video conferencing tool (e.g., Zoom). The input is a set of questions generated by the server and the user's answers. A voice generation AI on the device reads the questions aloud. The output is the data of the user's voice responses.

[1405] Step 6:

[1406] The server converts the user's voice response into text data in real time. The input is the user's voice response. The server converts the voice to text data using Google Cloud Speech-to-Text technology. The output is the text data response.

[1407] Step 7:

[1408] The server analyzes the text data responses using natural language processing technology. The input is the text data responses. The server analyzes the converted text and evaluates the specificity and logic of the responses. The output is evaluation data of the analysis results.

[1409] Step 8:

[1410] The server creates feedback materials based on the analysis results. The input is the evaluation data. The server generates feedback sentences using a Python script. The output is the feedback materials.

[1411] Step 9:

[1412] The server sends the created feedback materials to the interviewer. The input is the feedback materials. The server sends the materials via email. The output is a notification that the materials have been sent.

[1413] Step 10:

[1414] The server uses optical character recognition technology to extract robot status information from images. The input is image data showing the robot's status. The server uses OpenCV and pytesseract to convert the character information in the image into text data. The output is text data of the robot's status.

[1415] Step 11:

[1416] The server uses a generative artificial intelligence model to generate maintenance procedures tailored to specific situations. The input is text data of the robot's status. The server uses a generative AI model (e.g., GPT-3.5) to generate appropriate maintenance procedures. For example, the prompt sentence "The robot's status is displayed as 'Error message 72: Motor overheating'. Please suggest the next maintenance procedure." is input to the model. The output is the generated maintenance procedure.

[1417] Step 12:

[1418] The server evaluates the maintenance work based on the procedure. The input is the generated maintenance procedure and the maintenance work result data. The server evaluates the work performed according to the procedure and creates the results as feedback material. The output is the evaluation result of the maintenance work and the feedback material.

[1419] Step 13:

[1420] The server sends the created feedback material to the worker. The input is the feedback material. The server sends the material via email. The output is a notification that the material has been sent.

[1421] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1422] System Overview

[1423] This invention is a system that uses artificial intelligence (generative AI models) in the early stages of the hiring process. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate hiring process.

[1424] Resume analysis

[1425] The server receives resumes uploaded by candidates, converts the resume files (PDF or Word files) into text data using optical character recognition (OCR), and then uses natural language processing (NLP) to extract information such as educational background, work history, skills, and motivation for applying from the resume text.

[1426] Question Set Generation

[1427] The server generates a set of questions based on the extracted information, customized to include content relevant to the position being applied for. For example, a software engineer candidate might be asked questions like, "Tell me about your Python programming experience."

[1428] Conducting interviews

[1429] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). The candidate takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions. The user (candidate) answers them.

[1430] Incorporating an emotion engine

[1431] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is then sent to the server.

[1432] Analysis of response and sentiment data

[1433] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. The response is evaluated based on factors such as specificity and logic. Emotion data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed.

[1434] Knowledge test

[1435] The server generates knowledge test questions in a predetermined format and presents them to the candidate through the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[1436] Creating and Providing Feedback

[1437] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[1438] Specific examples of processing

[1439] For example, when a software engineer candidate uploads their resume, the server extracts the candidate's programming skills and project experience. During the interview, they are asked questions such as, "Tell us about your specific project experience using Python and Django," and the emotion engine analyzes the user's facial expressions and tone of voice while answering. The server analyzes the voice responses and emotion data and makes an evaluation such as, "The answers were highly specific and emotionally confident." This evaluation is reflected in the feedback materials and sent to the first interviewer.

[1440] By incorporating emotion recognition technology, this system enables objective evaluation that is not influenced by the interviewer's subjectivity, contributing to the selection of more appropriate personnel. It also improves convenience for applicants by providing flexible interview opportunities.

[1441] The processing flow will be explained below.

[1442] Step 1:

[1443] The server receives resume files uploaded by applicants, which are expected to be in PDF or Word format.

[1444] Step 2:

[1445] The server converts the resume file into text data using optical character recognition (OCR) technology, a process required for PDFs and scanned images.

[1446] Step 3:

[1447] The server uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data.

[1448] Step 4:

[1449] The server generates a set of questions based on the extracted information, including questions such as "Tell me about your Python programming experience."

[1450] Step 5:

[1451] The device will notify the candidate of the date, time and method of the interview (e.g., Zoom link) via email or messaging system.

[1452] Step 6:

[1453] The user will be interviewed at the appointed date and time using a video conferencing tool (Zoom), with the device reading out pre-generated questions using a speech-generating AI.

[1454] Step 7:

[1455] The user responds to questions posed by the voice-generating AI with their own voice. The device records the response and sends it to the server in real time.

[1456] Step 8:

[1457] During the interview, the device activates an emotion recognition engine that analyzes the user's facial expressions and tone of voice to extract emotional data in real time, which is then sent to the server.

[1458] Step 9:

[1459] The server uses speech recognition technology to convert the user's voice response into text data. The converted text is then analyzed using natural language processing technology to evaluate points such as the specificity and logic of the response. At the same time, emotional data obtained from an emotion recognition engine is also analyzed.

[1460] Step 10:

[1461] The server generates knowledge test questions in a predetermined format and presents them to candidates via their terminals. The questions can be multiple choice or short answer.

[1462] Step 11:

[1463] The user answers the questions in the knowledge test, and the terminal sends the answers to the server.

[1464] Step 12:

[1465] The server evaluates the candidate's knowledge test answers and calculates a score.

[1466] Step 13:

[1467] The server aggregates all responses and emotional data to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. The emotional data indicates the candidate's emotional changes during the interview and their overall emotional state.

[1468] Step 14:

[1469] The server transmits the created feedback material to the first interviewer, who can use this material to efficiently prepare for the next interview.

[1470] Example 2

[1471] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1472] In the traditional hiring process, it is easily influenced by the interviewer's subjectivity, making it difficult to make an objective evaluation. Furthermore, there is no way to evaluate the candidate's emotional changes and facial expressions in real time, making it difficult to make an accurate and fair evaluation. Furthermore, the typical interview process is time-consuming, labor-intensive, and inefficient.

[1473] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1474] In this invention, the server includes means for receiving and analyzing resumes, means for converting resume files into text data using optical character recognition technology, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for inputting prompts and generating appropriate questions using a generative AI model, means for conducting interviews using a video conferencing tool and reading out questions, means for converting voice responses into text and analyzing them using natural language processing technology, means for analyzing facial expressions and tone of voice during the interview using an emotion recognition engine and extracting emotion data, means for integrating the analysis results and the emotion data and evaluating them, means for creating feedback materials, and means for sending the feedback materials to interviewers. This enables objective and accurate evaluation, including changes in the emotions of candidates, and improves the efficiency and fairness of the hiring process.

[1475] "Means for receiving and analyzing resumes" refers to the device or software used to receive resume files uploaded by applicants and analyze their contents.

[1476] "Optical character recognition technology" is a technology that extracts text information from image data and is used to convert digital resume files into text data.

[1477] "Natural language processing technology" is a technology that uses computers to analyze and understand text written in natural language, and is used to extract necessary information from text data such as resumes.

[1478] A "means for generating a question set" is a device or software for generating appropriate interview questions based on the analyzed resume data.

[1479] A "generative AI model" is a trained artificial intelligence model that has the ability to generate natural language when given a prompt sentence.

[1480] A "prompt" is an instruction given to an AI model to perform a specific task.

[1481] A "video conferencing tool" is software for sharing audio and video in real time over the Internet.

[1482] A "means for converting voice responses to text" is a device or software that uses voice recognition technology to convert voice responses during an interview into text form.

[1483] An "emotion recognition engine" is a device or software that analyzes a person's emotional state from audio and video data and extracts emotional data in real time.

[1484] The "means for creating feedback materials" refers to a device or software for generating feedback materials summarizing the evaluation of candidates based on the results of the interview and knowledge test.

[1485] The "means for sending feedback materials to interviewers" refers to a device or software for automatically sending the created feedback materials to interviewers.

[1486] This invention is a system that combines artificial intelligence (generative AI models) and an emotion recognition engine to perform more accurate evaluations in the early stages of the hiring process. This system automates the processes of resume analysis, question generation, interviews, response analysis, knowledge testing, and feedback creation.

[1487] Hardware and software used

[1488] The system uses the following hardware and software:

[1489] 1. Server

[1490] A server with the storage and processing power to receive resumes

[1491] Optical Character Recognition (OCR) engine (e.g., Tesseract)

[1492] Natural Language Processing (NLP) engines (e.g., SpaCy, BERT)

[1493] Generative AI models (e.g., GPT-3)

[1494] Voice recognition technology (e.g., Google Speech-to-Text)

[1495] Report generation tools (e.g. Tableau)

[1496] 2. Terminal

[1497] Devices for real-time communication with users (PCs, smartphones, tablets, etc.)

[1498] Video conferencing tools (e.g., Zoom, Microsoft Teams)

[1499] Voice generation AI (e.g. Amazon Polly)

[1500] Emotion recognition engine (e.g. Microsoft Azure Emotion API)

[1501] 3. Users

[1502] Applicants and interviewers

[1503] Specific processing of the program

[1504] Resume upload and analysis

[1505] Users upload their resumes to the server, which converts the received resume files (PDF or Word files) into text data using an optical character recognition (OCR) engine. It then uses natural language processing (NLP) to extract information such as educational background, work history, and skills from the resume text data.

[1506] Question Set Generation

[1507] The server generates a set of relevant questions for the candidate based on the analyzed resume data. It uses a generative AI model (GPT-3) to generate appropriate questions. For example, the server inputs a prompt such as, "Generate interview questions based on the candidate's programming skills."

[1508] Preparing for and conducting interviews

[1509] The device notifies the candidate of the date, time, and method of the interview, sends a link to a video conferencing tool (such as Zoom), and the interview takes place at the specified date and time. During the interview, a speech-generation AI (Amazon Polly) is used to read out pre-generated questions.

[1510] Analysis by emotion engine

[1511] During the interview, the device runs an emotion recognition engine that analyzes the user's facial expressions and tone of voice, extracting emotional data in real time and sending it to the server.

[1512] Analysis of response and sentiment data

[1513] The server converts the applicant's voice responses into text data using speech recognition technology and analyzes them using natural language processing technology. The server evaluates the specificity and logic of the responses. It also analyzes emotional data obtained from an emotion recognition engine to evaluate emotional changes.

[1514] Knowledge test

[1515] The server creates knowledge test questions using a generative AI model and presents them to the candidate via their device. For example, the candidate might enter a prompt such as, "Please create a question about data structures and algorithms." The candidate's answers are then sent to the server for evaluation.

[1516] Creating and Providing Feedback

[1517] The server integrates the results of the interview and knowledge test to create feedback materials that include the candidate's strengths and weaknesses, as well as what needs to be checked next time. This feedback includes emotional data and is evaluated in detail. The created feedback materials are sent to the first interviewer.

[1518] Examples of prompt statements

[1519] 1. "Please parse this resume and extract Python and Django project experience."

[1520] 2. "Generate questions to assess knowledge of data structures and algorithms."

[1521] 3. "Evaluate the specificity of interview responses and create feedback materials that analyze emotional data."

[1522] By integrating emotion recognition and AI technology, the system eliminates subjective interviewer evaluations, significantly improving the accuracy and efficiency of the hiring process, while providing candidates with a flexible interview experience for increased convenience.

[1523] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1524] Step 1:

[1525] Resume upload and analysis

[1526] The user uploads their resume to the system. The server stores the received resume file (PDF or Word file) in storage. Optical character recognition (OCR) technology is used to convert the file into text data. The input is a PDF or Word file, and the output is text data. The server launches an OCR engine (e.g., Tesseract) to analyze the file and extract the text. Next, natural language processing (NLP) technology is used to extract information such as education, work history, and skills from the resume text data. An NLP engine (e.g., SpaCy) is used to identify the information for each section.

[1527] Step 2:

[1528] Question Set Generation

[1529] The server generates a set of questions relevant to the candidate based on the analyzed information. It uses a generative AI model (e.g., GPT-3) and inputs a prompt to generate questions. The input is the analyzed resume data and the prompt, and the output is a set of interview questions. For example, the prompt "Generate interview questions based on the candidate's programming skills" is input to the generative AI model. The generated questions are saved for interview preparation.

[1530] Step 3:

[1531] Preparing for and conducting interviews

[1532] The device notifies the candidate of the interview date, time, and method (e.g., Zoom link). Notification is sent via email or SMS. The input is the interview date, time, and method, and the output is the notification to the candidate. The user (candidate) takes the interview using a video conferencing tool at the notified date and time. During the interview, the device uses a speech generation AI (e.g., Amazon Polly) to read out pre-generated questions. The input is the generated question set, and the output is the audio reading.

[1533] Step 4:

[1534] Analysis by emotion engine

[1535] During the interview, the device runs an emotion recognition engine to analyze the user's (candidate's) facial expressions and tone of voice. Emotional data is extracted in real time and sent to the server. The input is video and audio data, and the output is emotion data. Specifically, a webcam and microphone are used to capture face and voice, and the emotion recognition engine (e.g., Microsoft Azure Emotion API) analyzes this data in real time.

[1536] Step 5:

[1537] Analysis of response and sentiment data

[1538] The server converts the candidate's voice response into text data using speech recognition technology (e.g., Google Speech-to-Text). It then analyzes the response using natural language processing technology. The input is the voice response and emotional data, and the output is the analyzed text and emotional evaluation. The specificity and logic of the response are evaluated, and data obtained from the emotion recognition engine is simultaneously analyzed to evaluate emotional changes. The text data is evaluated using an NLP engine (e.g., BERT).

[1539] Step 6:

[1540] Knowledge test

[1541] The server uses a generative AI model to create knowledge test questions and presents them to the candidate via their terminal. For example, a prompt such as "Please create questions about data structures and algorithms" is input to the generative AI model. The input is the prompt to the generative AI model, and the output is the knowledge test questions. The user answers the test, and the answers are sent back to the server. The input is the candidate's test answers, and the output is the evaluated answers.

[1542] Step 7:

[1543] Creating and Providing Feedback

[1544] The server integrates the results of the interview and knowledge test and creates feedback materials that include the candidate's strengths and weaknesses and what to check next time. The feedback also includes emotional data. The inputs are the interview results, knowledge test results, and emotional data, and the output is the feedback materials. The created feedback materials are sent to the first interviewer. Specifically, a report generation tool (e.g., Tableau) is used to provide feedback in a visually easy-to-understand format.

[1545] (Application example 2)

[1546] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1547] Traditional hiring processes and customer service in brick-and-mortar stores are prone to errors, time-consuming manual operations, and subjective evaluation biases by interviewers and store clerks. Furthermore, traditional systems are unable to recognize users' emotions in real time and respond accordingly. This creates a need for improved hiring efficiency and customer satisfaction.

[1548] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1549] In this invention, the server includes means for receiving and analyzing a resume, means for identifying sections of the resume using natural language processing technology, means for generating a question set based on the analysis results, means for conducting an interview using a video conferencing engine and reading out questions, means for converting voice responses to text and analyzing them using natural language processing technology, means for evaluating the analysis results and creating feedback materials, means for transmitting the feedback materials to the interviewer, an emotion recognition engine for extracting emotion data, means for analyzing the emotion data to evaluate the user's emotional state, means for answering the user's questions in real time, means for using natural language processing technology to understand the content of the user's questions, and means for analyzing and displaying data input to smart glasses or a head-mounted display. This enables efficient and objective evaluation in the recruitment process and customer service in physical stores, ultimately improving user satisfaction.

[1550] A "resume" is a document submitted by an applicant that includes information such as a self-introduction, educational background, work history, skills, and motivation for applying.

[1551] "Analysis" is the process of breaking down data or information to understand its content for a specific purpose.

[1552] "Natural language processing technology" is technology that enables computers to understand, interpret, and generate human language.

[1553] A "section" is a part of a document or data that is divided based on a specific theme or content.

[1554] A "question set" is a collection containing a series of questions created for a specific purpose.

[1555] A "video conferencing engine" is software or a system for visual and audio communication over the Internet.

[1556] A "voice response" is a response or reply given aloud.

[1557] "Convert to text" is the process of converting non-text data, such as audio or images, into character data.

[1558] "Feedback materials" are documents that compile information including the results of the evaluation and areas for improvement next time.

[1559] An "interviewer" is a person responsible for conducting an interview with an applicant and evaluating the results.

[1560] An "emotion recognition engine" is a system that detects and recognizes emotions from facial expressions, tone of voice, body movements, etc.

[1561] "Emotional state" refers to a person's emotional or psychological state at a given time.

[1562] "Smart glasses" are glasses with additional functions, such as displaying information and providing augmented reality functions.

[1563] A "head-mounted display" is a display device worn on the head to display images in front of the eyes.

[1564] System configuration

[1565] This invention is a system that combines artificial intelligence and emotion recognition technology for the early stages of the hiring process and customer service in brick-and-mortar stores. This system automates resume analysis, question generation, interviews, response analysis, and feedback creation, and by combining it with an emotion engine that recognizes and evaluates user emotions, it achieves a more accurate process. It can also be applied to real-time customer service in brick-and-mortar stores.

[1566] Resume analysis

[1567] The server receives resumes uploaded by applicants. It uses optical character recognition (OCR) technology to convert the resume file (PDF or Word file) into text data. It then uses natural language processing (NLP) technology to extract information such as educational background, work history, skills, and motivation for applying from the resume text data. Examples of software used here include Tesseract OCR and spaCy.

[1568] Question Set Generation

[1569] The server generates a set of questions based on the extracted information, customized to include content relevant to the position. For example, a generative AI model could be used to generate questions like, "Tell me about your Python programming experience."

[1570] Conducting interviews

[1571] The device notifies the applicant of the date, time, and method of the interview (for example, a Zoom link using a video conferencing engine). The applicant takes the interview at the notified date and time using a video conferencing tool. During the interview, a voice generation AI reads out pre-generated questions, and the user (applicant) answers them.

[1572] Incorporating an emotion engine

[1573] During the interview, the device runs an emotion recognition engine, which analyzes the user's facial expressions and tone of voice to extract emotional data in real time. This emotional data is sent to a server for analysis. For example, OpenCV or EmotionRecognizer is used.

[1574] Analysis of response and sentiment data

[1575] The server converts the user's voice response into text data using speech recognition technology and analyzes it using natural language processing technology. Points such as the specificity and logic of the response are also evaluated. Emotional data obtained from an emotion recognition engine is also used in the analysis to evaluate how the user's emotional state has changed. As a result, if the user's response is highly specific and emotionally confident, it can be evaluated as an "excellent answer."

[1576] Knowledge test

[1577] The server generates knowledge test questions in a predetermined format and presents them to the applicant via the terminal. The user answers the knowledge test questions and transmits the answer data to the server. The server evaluates these answers.

[1578] Creating and Providing Feedback

[1579] The server integrates the results of the interview and knowledge test to create feedback materials that include the applicant's strengths and weaknesses and what to check next time. This feedback also includes emotional data, showing detailed changes in emotions during the interview. The created feedback materials are sent to the first interviewer, who can use them to efficiently prepare for the next interview.

[1580] Application in physical stores

[1581] This system can also be applied to customer service in brick-and-mortar stores. For example, when a customer asks a question, a response can be made in real time. An emotion recognition engine analyzes the video input and evaluates customer satisfaction. This makes it possible to respond to customer needs quickly, which is expected to improve customer satisfaction.

[1582] Prompt Sentence Examples

[1583] "When a customer asks, 'Do you have this shirt in size large?', please provide an appropriate response based on your product inventory data."

[1584] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1585] Processing Steps of the Detailed Description

[1586] Step 1:

[1587] The server receives resumes uploaded by applicants. The submitted resumes are in PDF or Word file format and are converted into text data using optical character recognition (OCR) technology. Specifically, Tesseract OCR is used to analyze the characters in the image and generate text data.

[1588] Step 2:

[1589] The server analyzes the text data using natural language processing (NLP) technology to identify each section, such as educational background, work history, skills, motivation, etc. At this stage, the text is grammatically analyzed using, for example, spaCy, to extract the necessary information.

[1590] Step 3:

[1591] The server generates a set of questions based on the extracted information. Using a generative AI model, it generates questions relevant to the position being applied for. For example, specific questions such as "Tell me about your Python programming experience" are generated based on the prompt. This set of questions is output as text data.

[1592] Step 4:

[1593] The device notifies the applicant of the interview date and method (such as a Zoom link using a video conferencing engine). The notification is sent to the applicant in the form of a message or email. The user (applicant) receives the notification.

[1594] Step 5:

[1595] The user (applicant) is interviewed using a video conferencing tool. During the interview, the device uses a speech generation AI to read out pre-generated questions based on the question set generated in Step 3.

[1596] Step 6:

[1597] The user (applicant) answers questions by voice. The device collects the voice responses in real time and sends the voice data to the server.

[1598] Step 7:

[1599] The server converts the user's voice response into text data using voice recognition technology, such as the Google Speech-to-Text API.

[1600] Step 8:

[1601] The server then analyzes the converted text data using natural language processing technology to evaluate the specificity and logic of the answers. At this stage, text analysis is again performed using tools such as spaCy.

[1602] Step 9:

[1603] During the interview, the device runs an emotion recognition engine, analyzes the user's facial expressions and tone of voice, and extracts emotional data in real time. This emotional data is input as image and audio data and analyzed using OpenCV and EmotionRecognizer.

[1604] Step 10:

[1605] The server analyzes the emotional data and evaluates the user's emotional state. Specifically, it evaluates the user's emotional changes over time and quantifies their confidence and nervousness during the interview.

[1606] Step 11:

[1607] The server combines the analysis results of the interview voice responses with the emotional data to create feedback materials, which include the user's strengths and weaknesses and points to check next time. The feedback materials are output in text format.

[1608] Step 12:

[1609] The server sends the created feedback materials to the interviewer, who is then provided with the materials via email or a dedicated portal.

[1610] Step 13:

[1611] The server generates questions for the knowledge test in a predetermined format and presents them to the applicant via the terminal. The questions for the knowledge test are entered in text format, and the user answers them.

[1612] Step 14:

[1613] The user answers the knowledge test questions and sends the answer data to the server, which evaluates the knowledge test answers and reflects the results in the feedback material.

[1614] Step 15:

[1615] In a physical store, the device uses natural language processing technology and generative AI models to answer user questions in real time. For example, if a user asks, "Do you have this shirt in size L?", the device will provide an appropriate answer based on inventory data.

[1616] Step 16:

[1617] The device uses smart glasses or a head-mounted display to analyze the input data and provide visual information to the user. The user's visual data is analyzed in real time, and appropriate product information and guidance are displayed.

[1618] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1619] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1620] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1621] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1622] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1623] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1624] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1625] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1626] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1627] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1628] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1629] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1630] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1631] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1632] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1633] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1634] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1635] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1636] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1637] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1638] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1639] The following is further disclosed regarding the above embodiment.

[1640] (Claim 1)

[1641] A means of receiving and analyzing resumes;

[1642] a means for identifying each section of the resume using natural language processing techniques;

[1643] means for generating a question set based on the analysis results;

[1644] The interview will be conducted using a video conferencing tool, with questions read aloud,

[1645] A means for converting the voice response into text and analyzing it using natural language processing technology;

[1646] A means of evaluating the analysis results and generating feedback materials;

[1647] The system includes a means for sending feedback materials to interviewers.

[1648] (Claim 2)

[1649] means for administering a knowledge test in a predetermined format;

[1650] Includes a means of assessing knowledge test responses

[1651] 10. The system of claim 1.

[1652] (Claim 3)

[1653] Including a means to notify candidates of interview dates and methods

[1654] 10. The system of claim 1.

[1655] "Example 1"

[1656] (Claim 1)

[1657] A means of receiving and analyzing resumes;

[1658] means for converting the data into text data using optical character recognition technology;

[1659] A method for identifying each section of a resume using natural language processing technology and extracting information such as educational background, work history, skills, and motivation for applying;

[1660] means for generating a set of interview questions based on the extracted information;

[1661] A means to customize questions according to the company's requirements and the applicant's characteristics,

[1662] A means of informing candidates of the date and method of the interview;

[1663] Conducting interviews using video conferencing tools and reading questions aloud using voice generation technology;

[1664] means for converting the user's voice responses into text in real time;

[1665] The converted text is analyzed using natural language processing technology to evaluate the specificity and logic of the answers.

[1666] a means of generating feedback materials based on the results of the interviews and knowledge tests;

[1667] The system includes a means for sending feedback materials to interviewers.

[1668] (Claim 2)

[1669] means for generating and presenting knowledge test questions to candidates in a predetermined format;

[1670] 10. The system of claim 1, further comprising means for evaluating the knowledge test answers.

[1671] (Claim 3)

[1672] 10. The system of claim 1, further comprising means for notifying the candidate to attend an interview using a video conferencing tool at...

Claims

1. A means of receiving and analyzing resumes; a means for identifying each section of the resume using natural language processing techniques; means for generating a question set based on the analysis results; The interview will be conducted using a video conferencing tool, with questions read aloud, A means for converting the voice response into text and analyzing it using natural language processing technology; A means of evaluating the analysis results and generating feedback materials; The system includes a means for sending feedback materials to interviewers.

2. means for administering a knowledge test in a predetermined format; Includes a means of assessing knowledge test responses The system of claim 1 .

3. Including a means to notify candidates of interview dates and methods The system of claim 1 .

Images