system

Abstract

Provide a system. 【Solution means】 Means for obtaining past business history from a database, Means for converting the obtained business history into a data structure, Means for supplying the data structure to a machine learning model and training the model, Means for receiving an inquiry from a user, Means for analyzing the received inquiry using the machine learning model and generating an answer, Means for transmitting the generated answer to a user terminal, Means for aggregating user feedback on the received answer and using it to improve the machine learning model, A system including the above.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance as a response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Base station contract operations have multiple contract forms and complex business contents, and tend to rely on experienced personnel. For this reason, inefficiencies and personalization of the operations are progressing, making it difficult for young or new employees to efficiently perform their duties. Also, it is difficult to provide a consistent response based on past cases, and a standardized system for flexibly responding to inquiry contents is required.

Means for Solving the Problems

[0005] This invention provides a system that retrieves past business history from a database, converts it into a data structure, and then supplies it to a machine learning model for training. Upon receiving a user inquiry, the system analyzes it using the machine learning model and generates a rapid and consistent response. The response is sent to the user's terminal, and feedback from the user is aggregated to continuously improve the machine learning model, thereby achieving efficient and standardized business responses. This makes it possible to provide an environment where anyone can quickly perform contract work, regardless of their skills or experience.

[0006] "Past business history" refers to records of contracts and actions taken in the past, which are stored in a database.

[0007] "Data structure" refers to the organization and structuring of data, transforming it into a format suitable for information extraction and machine learning.

[0008] A "machine learning model" is a computational model composed of algorithms that automatically learn patterns and rules using data, and is used for prediction and analysis.

[0009] A "user" is an entity that uses the system to make inquiries or perform operations related to contractual matters, and usually refers to a business person or a customer.

[0010] An "inquiry" is a question or request sent by a user to a system, seeking specific information or actions.

[0011] "Analysis" is the process of applying a machine learning model to the content of an inquiry to derive the optimal answer or response.

[0012] A "response" is a system-generated response to a user's inquiry, containing necessary information and instructions.

[0013] "Feedback" refers to evaluations and opinions regarding the responses and results that users provide to the system, and is collected to help improve the system. [Brief explanation of the drawing]

[0014] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14]It is a sequence diagram showing the processing flow of a data processing system in Application Example 2 when a sentiment engine is combined.

Mode for Carrying Out the Invention

[0015] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0018] In the following embodiments, a numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0019] In the following embodiments, a numbered storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

[0023] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0024] As shown in Figure 1, the 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.

[0025] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and 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.

[0028] 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 perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0030] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0034] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0035] This invention is a system for streamlining and standardizing base station contract operations, utilizing past operational history and enabling the automatic generation of optimal responses to inquiries using machine learning models. Its specific operation is described below.

[0036] The server connects to the database and retrieves past contracts and correspondence history. This data is then formatted by the server into a format suitable for machine learning. The formatted data is then fed by the server to a machine learning model for training. The trained model learns patterns specific to contract work and has the ability to respond quickly to complex inquiries.

[0037] When a user enters a contract-related inquiry using their device, the device sends the inquiry to the server. The server passes the received inquiry to a machine learning model for analysis. As a result of this analysis, an answer optimized for contract operations is generated, and the server sends that answer to the device.

[0038] Users can view the answers provided on their devices. If the answers are unsatisfactory, users can provide feedback. This feedback is collected by the server and used to further improve the model.

[0039] As a concrete example, consider a case where a user asks, "Please tell me the appropriate contract process for installing a new base station." In this case, the server uses a model based on similar past cases to generate a response that includes the optimal process steps, and delivers it to the terminal. The user can then quickly proceed with the contract procedures based on these steps. Through this entire process, the system prevents contract work from becoming dependent on individual employees and enables consistent responses across the entire organization.

[0040] The following describes the processing flow.

[0041] Step 1:

[0042] The server connects to the database and retrieves past contract templates and correspondence history. This prepares the server to gather the necessary data.

[0043] Step 2:

[0044] The server converts the acquired data into a format suitable for machine learning. By removing noise and cleaning the data, a high-quality dataset is constructed.

[0045] Step 3:

[0046] The server supplies data to a machine learning model and trains the model. The server trains the model to learn patterns in base station contract operations.

[0047] Step 4:

[0048] The user enters their contract-related inquiry using a terminal. The user enters their specific question or request into the terminal and presses the send button.

[0049] Step 5:

[0050] The terminal sends the inquiry details to the server. The terminal formats the user's data appropriately and transfers it.

[0051] Step 6:

[0052] The server passes the received query to a machine learning model for analysis. The model generates the optimal response based on the trained data.

[0053] Step 7:

[0054] The server sends the generated response to the terminal. The server then sends the response back to the terminal in an appropriate format so that the user can easily understand it.

[0055] Step 8:

[0056] The terminal displays the answer to the user. This allows the user to confirm the answer to their inquiry and use it to their advantage in their work.

[0057] Step 9:

[0058] The user enters feedback on the answer into their device. This feedback may include expressions of satisfaction, suggestions for improvement, etc.

[0059] Step 10:

[0060] The device sends feedback to the server. The device formats the feedback correctly and passes it to the server.

[0061] Step 11:

[0062] The server collects feedback and uses it to improve the machine learning model. This improves the overall accuracy of the system, making future inquiries more accurate and faster.

[0063] (Example 1)

[0064] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0065] In traditional contract management, handling complex inquiries is often dependent on individual employees, resulting in insufficient efficiency and standardization of operations. Therefore, while consistency of information and prompt responses are required, obtaining uniform and optimal answers remains a challenge.

[0066] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0067] In this invention, the server includes means for acquiring past data history from an information storage device, means for converting the acquired data history into an information format, and means for analyzing received inquiries using a predictive analytics model and generating responses. This enables the streamlining and standardization of contract operations and provides quick and consistent answers even to complex inquiries.

[0068] "Past data history" refers to information such as contracts and correspondence records generated in the past during contract operations.

[0069] An "information storage device" is a system that organizes and stores data, and is used by users to retrieve information later.

[0070] "Information format" refers to data that has been transformed into a specific structure or format to be adapted for machine learning and analysis.

[0071] A "predictive analytics model" refers to an algorithm that uses machine learning techniques to learn patterns from past data and predict future inquiries and situations.

[0072] An "inquiry" refers to a question or request that a user enters into the system, seeking specific information or guidance on procedures.

[0073] A "response" is the result generated by a predictive analytics model based on its analysis, and includes specific information and instructions provided in response to a user's inquiry.

[0074] "Opinions" refer to feedback and evaluations of responses provided by users, and are used to improve the system.

[0075] "Denoising" refers to the process of removing unnecessary elements and inaccurate information from data, transforming it into a format suitable for analysis.

[0076] "Content organization" refers to processing and adjustments made to improve the consistency and clarity of data.

[0077] This invention is an advanced information processing system for streamlining and standardizing contract management. This enables rapid and accurate responses and prevents reliance on individual expertise in the work. The system primarily consists of the following components:

[0078] The server connects to an information storage device to retrieve historical data. A database management system is used for this, with PostgreSQL and MySQL (registered trademark) being typical examples. After retrieving the data, the server converts it into an information format using the pandas library. At this stage, processing is performed to remove noise and organize the content.

[0079] Next, the server feeds the formatted data to a predictive analytics model, which is then trained using scikit-learn and TENSORFLOW®. This model learns patterns specific to contract work and is then adapted to handle a variety of inquiries.

[0080] The user enters specific inquiries into the system via their terminal. For example, a prompt might read, "Please tell me the appropriate contract process for installing a new base station." This inquiry is then sent from the terminal to the server.

[0081] Subsequently, the server analyzes the query using a predictive analytics model and generates the optimal response. This response is based on similar past cases and provides consistent information to the user. The generated response is sent from the server to the terminal, where the user can review it.

[0082] If the response provided is unsatisfactory, users can provide feedback. This feedback is aggregated on the server and used to improve the model's performance. This allows the system to continuously improve, increasing the efficiency and consistency of contract management.

[0083] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0084] Step 1:

[0085] The server connects to the information storage device and retrieves historical data. The input is a contract-related dataset, and the data is extracted by executing SQL queries using a database management system. The output is a set of raw data. The goal of this step is to efficiently and accurately collect the necessary data.

[0086] Step 2:

[0087] This step converts the raw data acquired by the server into an informational format. The input is the raw data obtained in step 1, which is converted into a data frame using the pandas library, and then denoised and organized. The output is clean and structured data. Specific actions in this step include removing whitespace values ​​and normalizing text.

[0088] Step 3:

[0089] The server trains a predictive analytics model using the formatted data. The input is the formatted data from step 2, and the model is trained using libraries such as scikit-learn and TensorFlow. The output is the trained model. In this process, the model learns the patterns necessary for contract work.

[0090] Step 4:

[0091] The user enters specific prompt text on their terminal and sends an inquiry about the contract. The input is a prompt text, for example, "Please tell me the appropriate contract process for installing a new base station." The output is a structured data request sent to the server.

[0092] Step 5:

[0093] The server receives the prompt and performs analysis using a predictive analytics model. The input is the structured data request from step 4, and the model performs analysis using reference data. The output is an optimized response to the query. This analysis involves inference based on similar past cases.

[0094] Step 6:

[0095] The server sends the generated response to the user's terminal. The input is the response data generated in step 5, and the output is the information displayed on the user's terminal. Appropriate presentation of the response content is important here.

[0096] Step 7:

[0097] The user provides feedback on the response they receive and sends it from their terminal to the server. The input is the user's feedback data, and the output is aggregated information of the opinions on the server. This feedback contributes to improving the model and making decisions for the next steps.

[0098] (Application Example 1)

[0099] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0100] In traditional contract management, past business history could not be efficiently utilized, resulting in inefficient and person-dependent responses to various inquiries. Furthermore, even when there was a demand for improved response quality and faster response times, there was a lack of a system to properly manage and analyze the necessary data. As a result, providing immediate voice responses to user inquiries was difficult, and the system lacked flexibility and responsiveness.

[0101] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0102] In this invention, the server includes means for acquiring past history from an information management device, means for converting the acquired history into an information structure, means for supplying the information structure to a learning device and training the device, means for collecting voice input and converting voice into text information, and means for providing the generated response to the user as voice output. This makes it possible to respond to user inquiries quickly and accurately, thereby achieving efficiency and standardization of contract operations.

[0103] "Past history" refers to records of a series of operations and transactions that have taken place in the past, and is the information stored in the data management device.

[0104] An "information management device" is a device used for acquiring, storing, retrieving, and processing data, and includes database systems.

[0105] "Information structure" refers to the form of data that has been formatted in a way that can be handled by machine learning devices, and is a data format suitable for analysis and training.

[0106] A "learning device" is a computer system that uses machine learning algorithms to process data and perform analysis and predictions.

[0107] The term "user" refers to the entity that uses a system or service to perform contractual work or related tasks, and is usually a human operator or person in charge.

[0108] "Voice input" is an input method that electronically collects the voice spoken by the user and converts it into analyzable data.

[0109] "Converting audio to text information" is the process of converting collected audio data into text data using natural language processing technology.

[0110] "Voice output" refers to an output method for transmitting generated responses or information to the user in voice format, and speech synthesis technology supports this.

[0111] When implementing this invention, a high-performance server and a terminal such as a smartphone or smart glasses are required. In this embodiment, the system operates in the following steps.

[0112] The server first retrieves historical data from the information management device. This data is transformed into an information structure, and the data is formatted to minimize noise. This information structure is then supplied to a training device using a machine learning framework such as TensorFlow, and the device is trained. Through this training, the device acquires advanced analytical capabilities for queries.

[0113] Users make inquiries using voice input via smartphones or smart glasses. This voice is converted into text information within the device and sent to the server. The server uses a trained learning device to analyze the inquiry and generate the optimal response.

[0114] The generated response is sent back from the server to the user's terminal and provided to the user as voice output using speech synthesis technology. This entire process allows the user to quickly obtain information about contract-related questions and transaction details.

[0115] For example, when a user asks "Please let me know the payment due date for next month" through smart glasses, the system can calculate the payment timing from the history and inform the user via voice, "The next payment due date is May 15th."

[0116] An example of a prompt to a generative AI model would be, "Based on my recent payment history, please provide a detailed, step-by-step process for my next payment." By using this prompt, the AI ​​model can use past history to derive an accurate answer.

[0117] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0118] Step 1:

[0119] The server retrieves historical data from the information management device. At this stage, the input is past business data stored within the information management device, and all relevant information is passed to the server as output. The server analyzes the retrieved data and formats it into the format required for subsequent processing.

[0120] Step 2:

[0121] The server uses the formatted historical data to transform it into an information structure. The input here is the historical data obtained in step 1, and the output is a dataset that can be processed by a machine learning framework. Through data format transformation and refinement, clean data with noise removed is generated.

[0122] Step 3:

[0123] The server uses TensorFlow to supply information structures to the learning machine and train the learning model. The input in this step is the clean dataset prepared in step 2, and the output is a trained learning model suitable for query analysis. The model is optimized using advanced algorithms, leveraging the computing resources on the server side.

[0124] Step 4:

[0125] The user uses a smartphone or smart glasses to perform voice input. This input consists of the user's inquiry. The device converts the voice into text and sends it to the server. The output is the user's inquiry converted into text.

[0126] Step 5:

[0127] The server analyzes the received character information using a learning device and generates the optimal response. In this process, the input is the character information obtained in step 4, and the output is the generated response. Considering the prompt text and past history, the AI ​​model provides an accurate answer to the query.

[0128] Step 6:

[0129] The server sends the generated response to the terminal. The input is the response generated in step 5, and the output is the voice output delivered to the user using speech synthesis technology. The terminal uses the received response to provide real-time voice feedback to the user.

[0130] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0131] This invention relates to a system that not only improves the efficiency of base station contract operations but also enables more natural and optimal responses by recognizing user emotions. This system incorporates an emotion engine that utilizes past business history and enables responses that take user emotions into consideration.

[0132] The server retrieves past contract information and response history from the database, cleanses this data, and provides it to a machine learning model. Through this process, the model learns knowledge about contract operations and becomes able to generate highly accurate responses to different inquiries.

[0133] When a user submits an inquiry using a device, the device first sends the information to the emotion engine. The emotion engine analyzes the user's input text and voice to recognize their emotions. The server combines this emotion information to generate the most appropriate response based on the inquiry.

[0134] The generated response is sent from the server to the terminal and presented to the user. User feedback is also collected and used to improve both the sentiment engine and the machine learning model, further enhancing future responses.

[0135] As a concrete example, consider a scenario where a user expresses dissatisfaction, stating, "This contract procedure is extremely cumbersome." The emotion engine recognizes the user's dissatisfaction and determines that an urgent solution is needed. Based on this, the server presents simplified procedures or common solutions. In this way, utilizing an emotion engine makes it possible to improve the user experience.

[0136] The following describes the processing flow.

[0137] Step 1:

[0138] The server connects to the database and retrieves past contract information and service history. The server collects this data and prepares for the next processing step.

[0139] Step 2:

[0140] The server denoises and cleanses the acquired data, converting it into a format suitable for machine learning models. The server then stores this cleansed data.

[0141] Step 3:

[0142] The server supplies data to a machine learning model and trains the model. The server enables the model to learn contract patterns and generate accurate answers to inquiries.

[0143] Step 4:

[0144] The user enters and submits a contract-related inquiry using their device. The user clearly enters the necessary information and questions.

[0145] Step 5:

[0146] The device sends inquiry information to the emotion engine, which analyzes the user's emotional state. The emotion engine infers emotions from the user's text and voice.

[0147] Step 6:

[0148] The server receives emotion data and query content from the emotion engine, inputs the received information into a machine learning model for analysis, and generates a response adapted to the user's emotions.

[0149] Step 7:

[0150] The server sends the generated response to the terminal. The user receives the response through the terminal and confirms its contents.

[0151] Step 8:

[0152] Users enter feedback on their devices. This feedback includes their satisfaction with the response and suggestions for improvement.

[0153] Step 9:

[0154] The device sends feedback to the server. The server receives the feedback and uses it to improve the sentiment engine and machine learning models.

[0155] Step 10:

[0156] The server analyzes the feedback and adjusts the machine learning model and sentiment engine. This improves the accuracy and emotional response of the next user response.

[0157] (Example 2)

[0158] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0159] In traditional contract management, handling user inquiries was inefficient, and it was difficult to provide optimal responses while considering user emotions. Furthermore, there was insufficient mechanism for effectively utilizing user feedback to improve the system. As a result, the user experience was compromised, and operational productivity suffered.

[0160] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0161] In this invention, the server includes means for acquiring past business history from an information storage medium, means for converting the acquired business history into an information structure, means for supplying the information structure to a machine learning model and training the model, means for analyzing user input and recognizing emotional information, and means for generating an optimal response based on the emotional information. This makes it possible to generate natural and optimal responses that take the user's emotions into consideration, thereby improving the user experience.

[0162] An "information storage medium" is a medium from which data can be stored and retrieved as needed.

[0163] "Information structure" refers to the organization and arrangement of data into a format that is easily usable for a specific purpose.

[0164] A "machine learning model" is a computational model that learns from data and performs pattern recognition and classification.

[0165] "Emotional information" refers to information extracted from user input data to identify the state and type of emotions.

[0166] An "optimal response" is one that generates the most appropriate and effective response based on the user's inquiry and emotional information.

[0167] This invention is a system that improves the efficiency of contract work and the user experience by taking user emotions into consideration and providing optimal responses. Implementation primarily involves servers, terminals, and users.

[0168] The server retrieves past business history from the data storage medium. This data is converted into an information structure using data analysis software. This information structure is then fed into a generative AI model, which is trained to handle various queries. The software used includes data cleansing tools and machine learning libraries.

[0169] Users submit inquiries via text or voice using their devices. These inquiries are sent by the device to the sentiment engine. The sentiment engine uses natural language processing software to analyze the user's emotions. The emotion information is sent to a server, which uses machine learning models to generate the most appropriate response based on this information.

[0170] The generated answers are presented to the user via the device. User feedback is sent to the server and used to improve the entire system.

[0171] For example, if a user complains that "this contract procedure is very cumbersome," the emotion engine recognizes this dissatisfaction. The server then attempts to alleviate the user's dissatisfaction by providing simplified procedural information.

[0172] An example of a prompt to input into a generative AI model is, "Please tell me an effective way to respond when a user expresses dissatisfaction." Based on this prompt, the system will suggest an appropriate solution.

[0173] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0174] Step 1:

[0175] The server retrieves past business history from the information storage medium. Specifically, the server accesses the database using SQL queries and selects contract information and response history. The input is the business history in the database, and the output is the raw data imported into the server.

[0176] Step 2:

[0177] The server converts the acquired business history data into an information structure. Specifically, it performs actions such as converting the data format, deleting unnecessary fields, and filling in missing data. The input is the raw data acquired in step 1, and the output is the cleansed information structure.

[0178] Step 3:

[0179] The server supplies the cleansed information structure to the generating AI model and trains the model. Specifically, it formats the data appropriately and trains the model using machine learning libraries. The input is the information structure, and the output is the trained machine learning model.

[0180] Step 4:

[0181] The user sends an inquiry using a device. Specifically, the user inputs the inquiry via text or voice and presses the send button on the device. The input is the user's inquiry, and the output is the inquiry data sent from the device to the server.

[0182] Step 5:

[0183] The terminal sends inquiry data to the emotion engine. Specifically, the terminal performs a function that transfers the inquiry content to an analysis engine. The input is the user's inquiry data, and the output is the analysis result from the emotion engine.

[0184] Step 6:

[0185] The emotion engine analyzes user inquiries and recognizes emotional information. Specifically, it uses natural language processing to generate emotion vectors from input data. The input is user inquiry data, and the output is identified emotional information.

[0186] Step 7:

[0187] The server uses recognized sentiment information to leverage a machine learning model to generate the optimal response. Specifically, the server generates response candidates from the model and selects the one that best matches the sentiment. The input is sentiment information and query data, and the output is the optimized response.

[0188] Step 8:

[0189] The server sends the generated, optimized response to the terminal. Specifically, the server transmits the response to the terminal via network communication. The input is the response data generated within the server, and the output is the response displayed on the terminal.

[0190] Step 9:

[0191] Users review the answers presented on their devices and provide feedback. Specifically, users input satisfaction ratings and comments through a feedback form. The input is feedback data sent to the server, and the output is the collected user feedback.

[0192] Step 10:

[0193] The server improves the system based on the collected feedback. Specifically, it analyzes the feedback data and adjusts the emotion engine and machine learning models as needed. The input is user feedback data, and the output is improved model accuracy and corresponding techniques.

[0194] (Application Example 2)

[0195] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0196] In systems used within the home, a challenge is to appropriately understand user emotions and provide more natural and effective interactions. Current systems do not take user emotions into consideration, which can lead to a decrease in the quality of the user experience.

[0197] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0198] In this invention, the server includes means for obtaining past behavioral history from an information set, means for converting the obtained behavioral history into a data structure, and means for supplying the data structure to a learning algorithm and training the algorithm. This makes it possible to analyze the user's emotions and generate an optimal response that takes the analysis results into account.

[0199] "Past behavioral history" refers to records and data about the user's past actions and interactions.

[0200] An "information collection" is a collection of diverse information and data stored in a database or storage device.

[0201] A "data structure" is a way of organizing and arranging data in a specific format so that it can be processed efficiently by a computer.

[0202] A "learning algorithm" is a series of computational procedures that enable a machine to autonomously learn patterns and knowledge from data.

[0203] An "information request" is an action initiated by a user to ask questions or seek support.

[0204] An "output device" is a device or interface used to present information generated by a system to the user.

[0205] To realize this application, the system is constructed as follows: The server retrieves past behavioral history from an information set and converts it into a data structure. Next, this data structure is supplied to a learning algorithm to train the algorithm. As the learning algorithm, general machine learning frameworks and tools can be used. Specific software that can be used include TensorFlow and PyTorch.

[0206] The terminal receives information requests from the user and sends them to the server. The server analyzes the received information requests using a learning algorithm and generates the optimal response. In this process, an emotion recognition engine is also used to analyze the user's emotions. The emotion recognition engine detects the user's emotions from voice input and text input, and uses the analysis results to generate the optimal response.

[0207] The generated response is sent back to the terminal and presented to the user via an output device. This output device may be a robot's display or speaker. The user reacts to the provided response, and this feedback is collected again by the server. The server uses this feedback to improve its learning algorithm and emotion recognition engine, contributing to improved response accuracy in subsequent interactions.

[0208] For example, if a user requests information saying "I'm tired today," the server receives this and uses its emotion recognition engine to determine that the user is feeling tired. Based on this, it generates suggestions for relaxing music to play and presents them to the user through the device.

[0209] An example of a prompt message would be, "Please tell me how to identify emotions from user input and suggest appropriate relaxation methods if the user is fatigued."

[0210] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0211] Step 1:

[0212] The server retrieves past behavioral history from a set of information. Input data includes logs of past user interactions. The server organizes this data item by item and retrieves it from storage in a specific format. The output is a structured dataset of behavioral history.

[0213] Step 2:

[0214] The server converts the acquired behavioral history into a data structure. The input data is the behavioral history dataset created in step 1. The server converts this into a data frame or other data structure and performs denoising and cleaning processes. The output is a clean dataset suitable for the learning algorithm.

[0215] Step 3:

[0216] The server feeds the clean dataset obtained in the previous step to the learning algorithm and trains the algorithm. The input data is the dataset that was the output of step 2. The server uses TensorFlow or PyTorch to tune the parameters of the learning algorithm and build the model. The output is a trained model suitable for a specific task.

[0217] Step 4:

[0218] The terminal receives information requests from the user. Input data is direct input from the user in voice or text format. The terminal prepares this data to be sent to the server as text. Output is the information request data ready for transmission.

[0219] Step 5:

[0220] The server analyzes the received information request using a learning algorithm and generates the optimal response. The input data is the information request sent in step 4. The server analyzes the user's emotions using an emotion recognition engine and calculates the optimal response based on the analysis results. The output is the response data to be presented to the user.

[0221] Step 6:

[0222] The terminal presents the generated response to the user through an output device. The input data is the response data generated in step 5. The terminal performs operations to express the response to the user using a speaker or display. The output is the response in a format that the user can see or hear.

[0223] Step 7:

[0224] The user responds to the presented response. The input is the user's response and feedback. The terminal acquires this input and prepares it for analysis by sending it to the server. The output is the feedback data sent to the server.

[0225] Step 8:

[0226] The server uses feedback to improve the learning algorithm and emotion recognition engine. The input data is the feedback data acquired in step 7. The server analyzes the feedback and improves the system's response accuracy by retraining the learning model and adjusting its parameters. The output is the improved learning model and emotion recognition engine.

[0227] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0228] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0229] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0230] [Second Embodiment]

[0231] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0232] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0233] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0234] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0235] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0236] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0237] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0238] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0239] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0240] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0241] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0242] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0243] This invention is a system for streamlining and standardizing base station contract operations, utilizing past operational history and enabling the automatic generation of optimal responses to inquiries using machine learning models. Its specific operation is described below.

[0244] The server connects to the database and retrieves past contracts and correspondence history. This data is then formatted by the server into a format suitable for machine learning. The formatted data is then fed by the server to a machine learning model for training. The trained model learns patterns specific to contract work and has the ability to respond quickly to complex inquiries.

[0245] When a user enters a contract-related inquiry using their device, the device sends the inquiry to the server. The server passes the received inquiry to a machine learning model for analysis. As a result of this analysis, an answer optimized for contract operations is generated, and the server sends that answer to the device.

[0246] Users can view the answers provided on their devices. If the answers are unsatisfactory, users can provide feedback. This feedback is collected by the server and used to further improve the model.

[0247] As a concrete example, consider a case where a user asks, "Please tell me the appropriate contract process for installing a new base station." In this case, the server uses a model based on similar past cases to generate a response that includes the optimal process steps, and delivers it to the terminal. The user can then quickly proceed with the contract procedures based on these steps. Through this entire process, the system prevents contract work from becoming dependent on individual employees and enables consistent responses across the entire organization.

[0248] The following describes the processing flow.

[0249] Step 1:

[0250] The server connects to the database and retrieves past contract templates and correspondence history. This prepares the server to gather the necessary data.

[0251] Step 2:

[0252] The server converts the acquired data into a format suitable for machine learning. By removing noise and cleaning the data, a high-quality dataset is constructed.

[0253] Step 3:

[0254] The server supplies data to a machine learning model and trains the model. The server trains the model to learn patterns in base station contract operations.

[0255] Step 4:

[0256] The user enters their contract-related inquiry using a terminal. The user enters their specific question or request into the terminal and presses the send button.

[0257] Step 5:

[0258] The terminal sends the inquiry details to the server. The terminal formats the user's data appropriately and transfers it.

[0259] Step 6:

[0260] The server passes the received query to a machine learning model for analysis. The model generates the optimal response based on the trained data.

[0261] Step 7:

[0262] The server sends the generated response to the terminal. The server then sends the response back to the terminal in an appropriate format so that the user can easily understand it.

[0263] Step 8:

[0264] The terminal displays the answer to the user. This allows the user to confirm the answer to their inquiry and use it to their advantage in their work.

[0265] Step 9:

[0266] The user enters feedback on the answer into their device. This feedback may include expressions of satisfaction, suggestions for improvement, etc.

[0267] Step 10:

[0268] The device sends feedback to the server. The device formats the feedback correctly and passes it to the server.

[0269] Step 11:

[0270] The server collects feedback and uses it to improve the machine learning model. This improves the overall accuracy of the system, making future inquiries more accurate and faster.

[0271] (Example 1)

[0272] Next, we will describe Example 1. 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."

[0273] In traditional contract management, handling complex inquiries is often dependent on individual employees, resulting in insufficient efficiency and standardization of operations. Therefore, while consistency of information and prompt responses are required, obtaining uniform and optimal answers remains a challenge.

[0274] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0275] In this invention, the server includes means for acquiring past data history from an information storage device, means for converting the acquired data history into an information format, and means for analyzing received inquiries using a predictive analytics model and generating responses. This enables the streamlining and standardization of contract operations and provides quick and consistent answers even to complex inquiries.

[0276] "Past data history" refers to information such as contracts and correspondence records generated in the past during contract operations.

[0277] An "information storage device" is a system that organizes and stores data, and is used by users to retrieve information later.

[0278] "Information format" refers to data that has been converted into a specific structure or format for adaptation to machine learning or analysis.

[0279] "Predictive analysis model" refers to an algorithm that uses machine learning techniques to learn patterns from past data and is used to predict future inquiries or situations.

[0280] "Inquiry" refers to questions or requests input by a user to the system, seeking specific information or procedure guidelines.

[0281] "Response" is the result generated by the predictive analysis model based on analysis, and includes specific information and instructions provided in response to the user's inquiry.

[0282] "Opinion" refers to feedback or evaluation of the response provided by the user, and is utilized for system improvement.

[0283] "Noise removal" refers to the process of removing unnecessary elements or incorrect information from data and converting it into a state suitable for analysis.

[0284] "Content arrangement" refers to the processing and adjustment performed to improve the consistency and clarity of data. <�

[0285] The present invention is an advanced information processing system for streamlining and standardizing contract operations. This enables quick and accurate responses and prevents the personalization of operations. This system is mainly composed of the following means.

[0286] The server connects to an information storage device to obtain past data history. For this, a database management system is used, and PostgreSQL, MySQL, etc. are representative. After obtaining the data, the server converts it into information format using the pandas library. Here, processing for noise removal and content arrangement is performed.

[0287] Next, the server feeds the formatted data into a predictive analytics model, which is then trained using scikit-learn or TensorFlow. This model learns patterns specific to contract work and is then adapted to handle a variety of inquiries.

[0288] The user enters specific inquiries into the system via their terminal. For example, a prompt might read, "Please tell me the appropriate contract process for installing a new base station." This inquiry is then sent from the terminal to the server.

[0289] Subsequently, the server analyzes the query using a predictive analytics model and generates the optimal response. This response is based on similar past cases and provides consistent information to the user. The generated response is sent from the server to the terminal, where the user can review it.

[0290] If the response provided is unsatisfactory, users can provide feedback. This feedback is aggregated on the server and used to improve the model's performance. This allows the system to continuously improve, increasing the efficiency and consistency of contract management.

[0291] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0292] Step 1:

[0293] The server connects to the information storage device and retrieves historical data. The input is a contract-related dataset, and the data is extracted by executing SQL queries using a database management system. The output is a set of raw data. The goal of this step is to efficiently and accurately collect the necessary data.

[0294] Step 2:

[0295] This step converts the raw data acquired by the server into an informational format. The input is the raw data obtained in step 1, which is converted into a data frame using the pandas library, and then denoised and organized. The output is clean and structured data. Specific actions in this step include removing whitespace values ​​and normalizing text.

[0296] Step 3:

[0297] The server trains a predictive analytics model using the formatted data. The input is the formatted data from step 2, and the model is trained using libraries such as scikit-learn and TensorFlow. The output is the trained model. In this process, the model learns the patterns necessary for contract work.

[0298] Step 4:

[0299] The user enters specific prompt text on their terminal and sends an inquiry about the contract. The input is a prompt text, for example, "Please tell me the appropriate contract process for installing a new base station." The output is a structured data request sent to the server.

[0300] Step 5:

[0301] The server receives the prompt and performs analysis using a predictive analytics model. The input is the structured data request from step 4, and the model performs analysis using reference data. The output is an optimized response to the query. This analysis involves inference based on similar past cases.

[0302] Step 6:

[0303] The server sends the generated response to the user's terminal. The input is the response data generated in step 5, and the output is the information displayed on the user's terminal. Appropriate presentation of the response content is important here.

[0304] Step 7:

[0305] Feedback is input for the response provided by the user and sent from the terminal to the server. The input is the user's feedback data, and the output is the aggregated information of opinions at the server. This feedback contributes to improving the model and making decisions for the next step.

[0306] (Application Example 1)

[0307] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0308] In the conventional contract business, the past business history could not be utilized efficiently, and the responses to various inquiries were personal and inefficient. Also, even when improvement in the quality of responses and prompt response were required, there was a lack of a mechanism for appropriately managing and analyzing the necessary data. As a result, it was difficult to provide an immediate voice response to the user's inquiry, and there were problems with the flexibility and responsiveness of the system.

[0309] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following respective means.

[0310] In this invention, the server includes means for acquiring the past history from the information management device, means for converting the acquired history into an information structure, means for supplying the information structure to the learning device and training the device, means for collecting voice input and converting it from voice into character information, and means for providing the generated response to the user by voice output. Thereby, it becomes possible to respond quickly and accurately to the user's inquiry, and the efficiency and standardization of the contract business are achieved.

[0311] The "past history" refers to a record of a series of past operations and transactions, and indicates the information accumulated in the data management device.

[0312] An "information management device" is a device used for acquiring, storing, retrieving, and processing data, and includes database systems.

[0313] "Information structure" refers to the form of data that has been formatted in a way that can be handled by machine learning devices, and is a data format suitable for analysis and training.

[0314] A "learning device" is a computer system that uses machine learning algorithms to process data and perform analysis and predictions.

[0315] The term "user" refers to the entity that uses a system or service to perform contractual work or related tasks, and is usually a human operator or person in charge.

[0316] "Voice input" is an input method that electronically collects the voice spoken by the user and converts it into analyzable data.

[0317] "Converting audio to text information" is the process of converting collected audio data into text data using natural language processing technology.

[0318] "Voice output" refers to an output method for transmitting generated responses or information to the user in voice format, and speech synthesis technology supports this.

[0319] When implementing this invention, a high-performance server and a terminal such as a smartphone or smart glasses are required. In this embodiment, the system operates in the following steps.

[0320] The server first retrieves historical data from the information management device. This data is transformed into an information structure, and the data is formatted to minimize noise. This information structure is then supplied to a training device using a machine learning framework such as TensorFlow, and the device is trained. Through this training, the device acquires advanced analytical capabilities for queries.

[0321] Users make inquiries using voice input via smartphones or smart glasses. This voice is converted into text information within the device and sent to the server. The server uses a trained learning device to analyze the inquiry and generate the optimal response.

[0322] The generated response is sent back from the server to the user's terminal and provided to the user as voice output using speech synthesis technology. This entire process allows the user to quickly obtain information about contract-related questions and transaction details.

[0323] For example, when a user asks "Please let me know the payment due date for next month" through smart glasses, the system can calculate the payment timing from the history and inform the user via voice, "The next payment due date is May 15th."

[0324] An example of a prompt to a generative AI model would be, "Based on my recent payment history, please provide a detailed, step-by-step process for my next payment." By using this prompt, the AI ​​model can use past history to derive an accurate answer.

[0325] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0326] Step 1:

[0327] The server retrieves historical data from the information management device. At this stage, the input is past business data stored within the information management device, and all relevant information is passed to the server as output. The server analyzes the retrieved data and formats it into the format required for subsequent processing.

[0328] Step 2:

[0329] The server uses the formatted historical data to transform it into an information structure. The input here is the historical data obtained in step 1, and the output is a dataset that can be processed by a machine learning framework. Through data format transformation and refinement, clean data with noise removed is generated.

[0330] Step 3:

[0331] The server uses TensorFlow to supply information structures to the learning machine and train the learning model. The input in this step is the clean dataset prepared in step 2, and the output is a trained learning model suitable for query analysis. The model is optimized using advanced algorithms, leveraging the computing resources on the server side.

[0332] Step 4:

[0333] The user uses a smartphone or smart glasses to perform voice input. This input consists of the user's inquiry. The device converts the voice into text and sends it to the server. The output is the user's inquiry converted into text.

[0334] Step 5:

[0335] The server analyzes the received character information using a learning device and generates the optimal response. In this process, the input is the character information obtained in step 4, and the output is the generated response. Considering the prompt text and past history, the AI ​​model provides an accurate answer to the query.

[0336] Step 6:

[0337] The server sends the generated response to the terminal. The input is the response generated in step 5, and the output is the voice output delivered to the user using speech synthesis technology. The terminal uses the received response to provide real-time voice feedback to the user.

[0338] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0339] This invention relates to a system that not only improves the efficiency of base station contract operations but also enables more natural and optimal responses by recognizing user emotions. This system incorporates an emotion engine that utilizes past business history and enables responses that take user emotions into consideration.

[0340] The server retrieves past contract information and response history from the database, cleanses this data, and provides it to a machine learning model. Through this process, the model learns knowledge about contract operations and becomes able to generate highly accurate responses to different inquiries.

[0341] When a user submits an inquiry using a device, the device first sends the information to the emotion engine. The emotion engine analyzes the user's input text and voice to recognize their emotions. The server combines this emotion information to generate the most appropriate response based on the inquiry.

[0342] The generated response is sent from the server to the terminal and presented to the user. User feedback is also collected and used to improve both the sentiment engine and the machine learning model, further enhancing future responses.

[0343] As a concrete example, consider a scenario where a user expresses dissatisfaction, stating, "This contract procedure is extremely cumbersome." The emotion engine recognizes the user's dissatisfaction and determines that an urgent solution is needed. Based on this, the server presents simplified procedures or common solutions. In this way, utilizing an emotion engine makes it possible to improve the user experience.

[0344] The following describes the processing flow.

[0345] Step 1:

[0346] The server connects to the database and retrieves past contract information and service history. The server collects this data and prepares for the next processing step.

[0347] Step 2:

[0348] The server denoises and cleanses the acquired data, converting it into a format suitable for machine learning models. The server then stores this cleansed data.

[0349] Step 3:

[0350] The server supplies data to a machine learning model and trains the model. The server enables the model to learn contract patterns and generate accurate answers to inquiries.

[0351] Step 4:

[0352] The user enters and submits a contract-related inquiry using their device. The user clearly enters the necessary information and questions.

[0353] Step 5:

[0354] The device sends inquiry information to the emotion engine, which analyzes the user's emotional state. The emotion engine infers emotions from the user's text and voice.

[0355] Step 6:

[0356] The server receives emotion data and query content from the emotion engine, inputs the received information into a machine learning model for analysis, and generates a response adapted to the user's emotions.

[0357] Step 7:

[0358] The server sends the generated response to the terminal. The user receives the response through the terminal and confirms its contents.

[0359] Step 8:

[0360] Users enter feedback on their devices. This feedback includes their satisfaction with the response and suggestions for improvement.

[0361] Step 9:

[0362] The device sends feedback to the server. The server receives the feedback and uses it to improve the sentiment engine and machine learning models.

[0363] Step 10:

[0364] The server analyzes the feedback and adjusts the machine learning model and sentiment engine. This improves the accuracy and emotional response of the next user response.

[0365] (Example 2)

[0366] Next, we will describe Example 2. 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".

[0367] In traditional contract management, handling user inquiries was inefficient, and it was difficult to provide optimal responses while considering user emotions. Furthermore, there was insufficient mechanism for effectively utilizing user feedback to improve the system. As a result, the user experience was compromised, and operational productivity suffered.

[0368] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0369] In this invention, the server includes means for acquiring past business history from an information storage medium, means for converting the acquired business history into an information structure, means for supplying the information structure to a machine learning model and training the model, means for analyzing user input and recognizing emotional information, and means for generating an optimal response based on the emotional information. This makes it possible to generate natural and optimal responses that take the user's emotions into consideration, thereby improving the user experience.

[0370] An "information storage medium" is a medium from which data can be stored and retrieved as needed.

[0371] "Information structure" refers to the organization and arrangement of data into a format that is easily usable for a specific purpose.

[0372] A "machine learning model" is a computational model that learns from data and performs pattern recognition and classification.

[0373] "Emotional information" refers to information extracted from user input data to identify the state and type of emotions.

[0374] An "optimal response" is one that generates the most appropriate and effective response based on the user's inquiry and emotional information.

[0375] This invention is a system that improves the efficiency of contract work and the user experience by taking user emotions into consideration and providing optimal responses. Implementation primarily involves servers, terminals, and users.

[0376] The server retrieves past business history from the data storage medium. This data is converted into an information structure using data analysis software. This information structure is then fed into a generative AI model, which is trained to handle various queries. The software used includes data cleansing tools and machine learning libraries.

[0377] Users submit inquiries via text or voice using their devices. These inquiries are sent by the device to the sentiment engine. The sentiment engine uses natural language processing software to analyze the user's emotions. The emotion information is sent to a server, which uses machine learning models to generate the most appropriate response based on this information.

[0378] The generated answers are presented to the user via the device. User feedback is sent to the server and used to improve the entire system.

[0379] For example, if a user complains that "this contract procedure is very cumbersome," the emotion engine recognizes this dissatisfaction. The server then attempts to alleviate the user's dissatisfaction by providing simplified procedural information.

[0380] An example of a prompt to input into a generative AI model is, "Please tell me an effective way to respond when a user expresses dissatisfaction." Based on this prompt, the system will suggest an appropriate solution.

[0381] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0382] Step 1:

[0383] The server retrieves past business history from the information storage medium. Specifically, the server accesses the database using SQL queries and selects contract information and response history. The input is the business history in the database, and the output is the raw data imported into the server.

[0384] Step 2:

[0385] The server converts the acquired business history data into an information structure. Specifically, it performs actions such as converting the data format, deleting unnecessary fields, and filling in missing data. The input is the raw data acquired in step 1, and the output is the cleansed information structure.

[0386] Step 3:

[0387] The server supplies the cleansed information structure to the generating AI model and trains the model. Specifically, it formats the data appropriately and trains the model using machine learning libraries. The input is the information structure, and the output is the trained machine learning model.

[0388] Step 4:

[0389] The user sends an inquiry using a device. Specifically, the user inputs the inquiry via text or voice and presses the send button on the device. The input is the user's inquiry, and the output is the inquiry data sent from the device to the server.

[0390] Step 5:

[0391] The terminal sends inquiry data to the emotion engine. Specifically, the terminal performs a function that transfers the inquiry content to an analysis engine. The input is the user's inquiry data, and the output is the analysis result from the emotion engine.

[0392] Step 6:

[0393] The emotion engine analyzes user inquiries and recognizes emotional information. Specifically, it uses natural language processing to generate emotion vectors from input data. The input is user inquiry data, and the output is identified emotional information.

[0394] Step 7:

[0395] The server uses recognized sentiment information to leverage a machine learning model to generate the optimal response. Specifically, the server generates response candidates from the model and selects the one that best matches the sentiment. The input is sentiment information and query data, and the output is the optimized response.

[0396] Step 8:

[0397] The server sends the generated, optimized response to the terminal. Specifically, the server transmits the response to the terminal via network communication. The input is the response data generated within the server, and the output is the response displayed on the terminal.

[0398] Step 9:

[0399] Users review the answers presented on their devices and provide feedback. Specifically, users input satisfaction ratings and comments through a feedback form. The input is feedback data sent to the server, and the output is the collected user feedback.

[0400] Step 10:

[0401] The server improves the system based on the collected feedback. Specifically, it analyzes the feedback data and adjusts the emotion engine and machine learning models as needed. The input is user feedback data, and the output is improved model accuracy and corresponding techniques.

[0402] (Application Example 2)

[0403] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the smart glasses 214 as the "terminal".

[0404] In systems used within the home, a challenge is to appropriately understand user emotions and provide more natural and effective interactions. Current systems do not take user emotions into consideration, which can lead to a decrease in the quality of the user experience.

[0405] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0406] In this invention, the server includes means for obtaining past behavioral history from an information set, means for converting the obtained behavioral history into a data structure, and means for supplying the data structure to a learning algorithm and training the algorithm. This makes it possible to analyze the user's emotions and generate an optimal response that takes the analysis results into account.

[0407] "Past behavioral history" refers to records and data about the user's past actions and interactions.

[0408] An "information collection" is a collection of diverse information and data stored in a database or storage device.

[0409] A "data structure" is a way of organizing and arranging data in a specific format so that it can be processed efficiently by a computer.

[0410] A "learning algorithm" is a series of computational procedures that enable a machine to autonomously learn patterns and knowledge from data.

[0411] An "information request" is an action initiated by a user to ask questions or seek support.

[0412] An "output device" is a device or interface used to present information generated by a system to the user.

[0413] To realize this application, the system is constructed as follows: The server retrieves past behavioral history from an information set and converts it into a data structure. Next, this data structure is supplied to a learning algorithm to train the algorithm. As the learning algorithm, general machine learning frameworks and tools can be used. Specific software that can be used include TensorFlow and PyTorch.

[0414] The terminal receives information requests from the user and sends them to the server. The server analyzes the received information requests using a learning algorithm and generates the optimal response. In this process, an emotion recognition engine is also used to analyze the user's emotions. The emotion recognition engine detects the user's emotions from voice input and text input, and uses the analysis results to generate the optimal response.

[0415] The generated response is sent back to the terminal and presented to the user via an output device. This output device may be a robot's display or speaker. The user reacts to the provided response, and this feedback is collected again by the server. The server uses this feedback to improve its learning algorithm and emotion recognition engine, contributing to improved response accuracy in subsequent interactions.

[0416] For example, if a user requests information saying "I'm tired today," the server receives this and uses its emotion recognition engine to determine that the user is feeling tired. Based on this, it generates suggestions for relaxing music to play and presents them to the user through the device.

[0417] An example of a prompt message would be, "Please tell me how to identify emotions from user input and suggest appropriate relaxation methods if the user is fatigued."

[0418] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0419] Step 1:

[0420] The server retrieves past behavioral history from a set of information. Input data includes logs of past user interactions. The server organizes this data item by item and retrieves it from storage in a specific format. The output is a structured dataset of behavioral history.

[0421] Step 2:

[0422] The server converts the acquired behavioral history into a data structure. The input data is the behavioral history dataset created in step 1. The server converts this into a data frame or other data structure and performs denoising and cleaning processes. The output is a clean dataset suitable for the learning algorithm.

[0423] Step 3:

[0424] The server feeds the clean dataset obtained in the previous step to the learning algorithm and trains the algorithm. The input data is the dataset that was the output of step 2. The server uses TensorFlow or PyTorch to tune the parameters of the learning algorithm and build the model. The output is a trained model suitable for a specific task.

[0425] Step 4:

[0426] The terminal receives information requests from the user. Input data is direct input from the user in voice or text format. The terminal prepares this data to be sent to the server as text. Output is the information request data ready for transmission.

[0427] Step 5:

[0428] The server analyzes the received information request using a learning algorithm and generates the optimal response. The input data is the information request sent in step 4. The server analyzes the user's emotions using an emotion recognition engine and calculates the optimal response based on the analysis results. The output is the response data to be presented to the user.

[0429] Step 6:

[0430] The terminal presents the generated response to the user through an output device. The input data is the response data generated in step 5. The terminal performs operations to express the response to the user using a speaker or display. The output is the response in a format that the user can see or hear.

[0431] Step 7:

[0432] The user responds to the presented response. The input is the user's response and feedback. The terminal acquires this input and prepares it for analysis by sending it to the server. The output is the feedback data sent to the server.

[0433] Step 8:

[0434] The server uses feedback to improve the learning algorithm and emotion recognition engine. The input data is the feedback data acquired in step 7. The server analyzes the feedback and improves the system's response accuracy by retraining the learning model and adjusting its parameters. The output is the improved learning model and emotion recognition engine.

[0435] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0436] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0437] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0438] [Third Embodiment]

[0439] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0440] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0441] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0442] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0443] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0444] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0445] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0446] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0447] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0448] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0449] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0450] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0451] This invention is a system for streamlining and standardizing base station contract operations, utilizing past operational history and enabling the automatic generation of optimal responses to inquiries using machine learning models. Its specific operation is described below.

[0452] The server connects to the database and retrieves past contracts and correspondence history. This data is then formatted by the server into a format suitable for machine learning. The formatted data is then fed by the server to a machine learning model for training. The trained model learns patterns specific to contract work and has the ability to respond quickly to complex inquiries.

[0453] When a user enters a contract-related inquiry using their device, the device sends the inquiry to the server. The server passes the received inquiry to a machine learning model for analysis. As a result of this analysis, an answer optimized for contract operations is generated, and the server sends that answer to the device.

[0454] Users can view the answers provided on their devices. If the answers are unsatisfactory, users can provide feedback. This feedback is collected by the server and used to further improve the model.

[0455] As a concrete example, consider a case where a user asks, "Please tell me the appropriate contract process for installing a new base station." In this case, the server uses a model based on similar past cases to generate a response that includes the optimal process steps, and delivers it to the terminal. The user can then quickly proceed with the contract procedures based on these steps. Through this entire process, the system prevents contract work from becoming dependent on individual employees and enables consistent responses across the entire organization.

[0456] The following describes the processing flow.

[0457] Step 1:

[0458] The server connects to the database and retrieves past contract templates and correspondence history. This prepares the server to gather the necessary data.

[0459] Step 2:

[0460] The server converts the acquired data into a format suitable for machine learning. By removing noise and cleaning the data, a high-quality dataset is constructed.

[0461] Step 3:

[0462] The server supplies data to a machine learning model and trains the model. The server trains the model to learn patterns in base station contract operations.

[0463] Step 4:

[0464] The user enters their contract-related inquiry using a terminal. The user enters their specific question or request into the terminal and presses the send button.

[0465] Step 5:

[0466] The terminal sends the inquiry details to the server. The terminal formats the user's data appropriately and transfers it.

[0467] Step 6:

[0468] The server passes the received query to a machine learning model for analysis. The model generates the optimal response based on the trained data.

[0469] Step 7:

[0470] The server sends the generated response to the terminal. The server then sends the response back to the terminal in an appropriate format so that the user can easily understand it.

[0471] Step 8:

[0472] The terminal displays the answer to the user. This allows the user to confirm the answer to their inquiry and use it to their advantage in their work.

[0473] Step 9:

[0474] The user enters feedback on the answer into their device. This feedback may include expressions of satisfaction, suggestions for improvement, etc.

[0475] Step 10:

[0476] The device sends feedback to the server. The device formats the feedback correctly and passes it to the server.

[0477] Step 11:

[0478] The server collects feedback and uses it to improve the machine learning model. This improves the overall accuracy of the system, making future inquiries more accurate and faster.

[0479] (Example 1)

[0480] Next, we will describe Example 1. 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."

[0481] In traditional contract management, handling complex inquiries is often dependent on individual employees, resulting in insufficient efficiency and standardization of operations. Therefore, while consistency of information and prompt responses are required, obtaining uniform and optimal answers remains a challenge.

[0482] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0483] In this invention, the server includes means for acquiring past data history from an information storage device, means for converting the acquired data history into an information format, and means for analyzing received inquiries using a predictive analytics model and generating responses. This enables the streamlining and standardization of contract operations and provides quick and consistent answers even to complex inquiries.

[0484] "Past data history" refers to information such as contracts and correspondence records generated in the past during contract operations.

[0485] An "information storage device" is a system that organizes and stores data, and is used by users to retrieve information later.

[0486] "Information format" refers to data that has been transformed into a specific structure or format to be adapted for machine learning and analysis.

[0487] A "predictive analytics model" refers to an algorithm that uses machine learning techniques to learn patterns from past data and predict future inquiries and situations.

[0488] An "inquiry" refers to a question or request that a user enters into the system, seeking specific information or guidance on procedures.

[0489] A "response" is the result generated by a predictive analytics model based on its analysis, and includes specific information and instructions provided in response to a user's inquiry.

[0490] "Opinions" refer to feedback and evaluations of responses provided by users, and are used to improve the system.

[0491] "Denoising" refers to the process of removing unnecessary elements and inaccurate information from data, transforming it into a format suitable for analysis.

[0492] "Content organization" refers to processing and adjustments made to improve the consistency and clarity of data.

[0493] This invention is an advanced information processing system for streamlining and standardizing contract management. This enables rapid and accurate responses and prevents reliance on individual expertise in the work. The system primarily consists of the following components:

[0494] The server connects to an information storage device to retrieve historical data. A database management system is used for this, with PostgreSQL and MySQL being typical examples. After retrieving the data, the server converts it into an information format using the pandas library. At this stage, processing is performed to remove noise and organize the content.

[0495] Next, the server feeds the formatted data into a predictive analytics model, which is then trained using scikit-learn or TensorFlow. This model learns patterns specific to contract work and is then adapted to handle a variety of inquiries.

[0496] The user enters specific inquiries into the system via their terminal. For example, a prompt might read, "Please tell me the appropriate contract process for installing a new base station." This inquiry is then sent from the terminal to the server.

[0497] Subsequently, the server analyzes the query using a predictive analytics model and generates the optimal response. This response is based on similar past cases and provides consistent information to the user. The generated response is sent from the server to the terminal, where the user can review it.

[0498] If the response provided is unsatisfactory, users can provide feedback. This feedback is aggregated on the server and used to improve the model's performance. This allows the system to continuously improve, increasing the efficiency and consistency of contract management.

[0499] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0500] Step 1:

[0501] The server connects to the information storage device and retrieves historical data. The input is a contract-related dataset, and the data is extracted by executing SQL queries using a database management system. The output is a set of raw data. The goal of this step is to efficiently and accurately collect the necessary data.

[0502] Step 2:

[0503] This step converts the raw data acquired by the server into an informational format. The input is the raw data obtained in step 1, which is converted into a data frame using the pandas library, and then denoised and organized. The output is clean and structured data. Specific actions in this step include removing whitespace values ​​and normalizing text.

[0504] Step 3:

[0505] The server trains a predictive analytics model using the formatted data. The input is the formatted data from step 2, and the model is trained using libraries such as scikit-learn and TensorFlow. The output is the trained model. In this process, the model learns the patterns necessary for contract work.

[0506] Step 4:

[0507] The user enters specific prompt text on their terminal and sends an inquiry about the contract. The input is a prompt text, for example, "Please tell me the appropriate contract process for installing a new base station." The output is a structured data request sent to the server.

[0508] Step 5:

[0509] The server receives the prompt and performs analysis using a predictive analytics model. The input is the structured data request from step 4, and the model performs analysis using reference data. The output is an optimized response to the query. This analysis involves inference based on similar past cases.

[0510] Step 6:

[0511] The server sends the generated response to the user's terminal. The input is the response data generated in step 5, and the output is the information displayed on the user's terminal. Appropriate presentation of the response content is important here.

[0512] Step 7:

[0513] The user provides feedback on the response they receive and sends it from their terminal to the server. The input is the user's feedback data, and the output is aggregated information of the opinions on the server. This feedback contributes to improving the model and making decisions for the next steps.

[0514] (Application Example 1)

[0515] Next, we will explain Application Example 1. In the following explanation, 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."

[0516] In traditional contract management, past business history could not be efficiently utilized, resulting in inefficient and person-dependent responses to various inquiries. Furthermore, even when there was a demand for improved response quality and faster response times, there was a lack of a system to properly manage and analyze the necessary data. As a result, providing immediate voice responses to user inquiries was difficult, and the system lacked flexibility and responsiveness.

[0517] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0518] In this invention, the server includes means for acquiring past history from an information management device, means for converting the acquired history into an information structure, means for supplying the information structure to a learning device and training the device, means for collecting voice input and converting voice into text information, and means for providing the generated response to the user as voice output. This makes it possible to respond to user inquiries quickly and accurately, thereby achieving efficiency and standardization of contract operations.

[0519] "Past history" refers to records of a series of operations and transactions that have taken place in the past, and is the information stored in the data management device.

[0520] An "information management device" is a device used for acquiring, storing, retrieving, and processing data, and includes database systems.

[0521] "Information structure" refers to the form of data that has been formatted in a way that can be handled by machine learning devices, and is a data format suitable for analysis and training.

[0522] A "learning device" is a computer system that uses machine learning algorithms to process data and perform analysis and predictions.

[0523] The term "user" refers to the entity that uses a system or service to perform contractual work or related tasks, and is usually a human operator or person in charge.

[0524] "Voice input" is an input method that electronically collects the voice spoken by the user and converts it into analyzable data.

[0525] "Converting audio to text information" is the process of converting collected audio data into text data using natural language processing technology.

[0526] "Voice output" refers to an output method for transmitting generated responses or information to the user in voice format, and speech synthesis technology supports this.

[0527] When implementing this invention, a high-performance server and a terminal such as a smartphone or smart glasses are required. In this embodiment, the system operates in the following steps.

[0528] The server first retrieves historical data from the information management device. This data is transformed into an information structure, and the data is formatted to minimize noise. This information structure is then supplied to a training device using a machine learning framework such as TensorFlow, and the device is trained. Through this training, the device acquires advanced analytical capabilities for queries.

[0529] Users make inquiries using voice input via smartphones or smart glasses. This voice is converted into text information within the device and sent to the server. The server uses a trained learning device to analyze the inquiry and generate the optimal response.

[0530] The generated response is sent back from the server to the user's terminal and provided to the user as voice output using speech synthesis technology. This entire process allows the user to quickly obtain information about contract-related questions and transaction details.

[0531] For example, when a user asks "Please let me know the payment due date for next month" through smart glasses, the system can calculate the payment timing from the history and inform the user via voice, "The next payment due date is May 15th."

[0532] An example of a prompt to a generative AI model would be, "Based on my recent payment history, please provide a detailed, step-by-step process for my next payment." By using this prompt, the AI ​​model can use past history to derive an accurate answer.

[0533] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0534] Step 1:

[0535] The server retrieves historical data from the information management device. At this stage, the input is past business data stored within the information management device, and all relevant information is passed to the server as output. The server analyzes the retrieved data and formats it into the format required for subsequent processing.

[0536] Step 2:

[0537] The server uses the formatted historical data to transform it into an information structure. The input here is the historical data obtained in step 1, and the output is a dataset that can be processed by a machine learning framework. Through data format transformation and refinement, clean data with noise removed is generated.

[0538] Step 3:

[0539] The server uses TensorFlow to supply information structures to the learning machine and train the learning model. The input in this step is the clean dataset prepared in step 2, and the output is a trained learning model suitable for query analysis. The model is optimized using advanced algorithms, leveraging the computing resources on the server side.

[0540] Step 4:

[0541] The user uses a smartphone or smart glasses to perform voice input. This input consists of the user's inquiry. The device converts the voice into text and sends it to the server. The output is the user's inquiry converted into text.

[0542] Step 5:

[0543] The server analyzes the received character information using a learning device and generates the optimal response. In this process, the input is the character information obtained in step 4, and the output is the generated response. Considering the prompt text and past history, the AI ​​model provides an accurate answer to the query.

[0544] Step 6:

[0545] The server sends the generated response to the terminal. The input is the response generated in step 5, and the output is the voice output delivered to the user using speech synthesis technology. The terminal uses the received response to provide real-time voice feedback to the user.

[0546] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0547] This invention relates to a system that not only improves the efficiency of base station contract operations but also enables more natural and optimal responses by recognizing user emotions. This system incorporates an emotion engine that utilizes past business history and enables responses that take user emotions into consideration.

[0548] The server retrieves past contract information and response history from the database, cleanses this data, and provides it to a machine learning model. Through this process, the model learns knowledge about contract operations and becomes able to generate highly accurate responses to different inquiries.

[0549] When a user submits an inquiry using a device, the device first sends the information to the emotion engine. The emotion engine analyzes the user's input text and voice to recognize their emotions. The server combines this emotion information to generate the most appropriate response based on the inquiry.

[0550] The generated response is sent from the server to the terminal and presented to the user. User feedback is also collected and used to improve both the sentiment engine and the machine learning model, further enhancing future responses.

[0551] As a concrete example, consider a scenario where a user expresses dissatisfaction, stating, "This contract procedure is extremely cumbersome." The emotion engine recognizes the user's dissatisfaction and determines that an urgent solution is needed. Based on this, the server presents simplified procedures or common solutions. In this way, utilizing an emotion engine makes it possible to improve the user experience.

[0552] The following describes the processing flow.

[0553] Step 1:

[0554] The server connects to the database and retrieves past contract information and service history. The server collects this data and prepares for the next processing step.

[0555] Step 2:

[0556] The server denoises and cleanses the acquired data, converting it into a format suitable for machine learning models. The server then stores this cleansed data.

[0557] Step 3:

[0558] The server supplies data to a machine learning model and trains the model. The server enables the model to learn contract patterns and generate accurate answers to inquiries.

[0559] Step 4:

[0560] The user enters and submits a contract-related inquiry using their device. The user clearly enters the necessary information and questions.

[0561] Step 5:

[0562] The device sends inquiry information to the emotion engine, which analyzes the user's emotional state. The emotion engine infers emotions from the user's text and voice.

[0563] Step 6:

[0564] The server receives emotion data and query content from the emotion engine, inputs the received information into a machine learning model for analysis, and generates a response adapted to the user's emotions.

[0565] Step 7:

[0566] The server sends the generated response to the terminal. The user receives the response through the terminal and confirms its contents.

[0567] Step 8:

[0568] Users enter feedback on their devices. This feedback includes their satisfaction with the response and suggestions for improvement.

[0569] Step 9:

[0570] The device sends feedback to the server. The server receives the feedback and uses it to improve the sentiment engine and machine learning models.

[0571] Step 10:

[0572] The server analyzes the feedback and adjusts the machine learning model and sentiment engine. This improves the accuracy and emotional response of the next user response.

[0573] (Example 2)

[0574] Next, we will describe Example 2. 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."

[0575] In traditional contract management, handling user inquiries was inefficient, and it was difficult to provide optimal responses while considering user emotions. Furthermore, there was insufficient mechanism for effectively utilizing user feedback to improve the system. As a result, the user experience was compromised, and operational productivity suffered.

[0576] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0577] In this invention, the server includes means for acquiring past business history from an information storage medium, means for converting the acquired business history into an information structure, means for supplying the information structure to a machine learning model and training the model, means for analyzing user input and recognizing emotional information, and means for generating an optimal response based on the emotional information. This makes it possible to generate natural and optimal responses that take the user's emotions into consideration, thereby improving the user experience.

[0578] An "information storage medium" is a medium from which data can be stored and retrieved as needed.

[0579] "Information structure" refers to the organization and arrangement of data into a format that is easily usable for a specific purpose.

[0580] A "machine learning model" is a computational model that learns from data and performs pattern recognition and classification.

[0581] "Emotional information" refers to information extracted from user input data to identify the state and type of emotions.

[0582] An "optimal response" is one that generates the most appropriate and effective response based on the user's inquiry and emotional information.

[0583] This invention is a system that improves the efficiency of contract work and the user experience by taking user emotions into consideration and providing optimal responses. Implementation primarily involves servers, terminals, and users.

[0584] The server retrieves past business history from the data storage medium. This data is converted into an information structure using data analysis software. This information structure is then fed into a generative AI model, which is trained to handle various queries. The software used includes data cleansing tools and machine learning libraries.

[0585] Users submit inquiries via text or voice using their devices. These inquiries are sent by the device to the sentiment engine. The sentiment engine uses natural language processing software to analyze the user's emotions. The emotion information is sent to a server, which uses machine learning models to generate the most appropriate response based on this information.

[0586] The generated answers are presented to the user via the device. User feedback is sent to the server and used to improve the entire system.

[0587] For example, if a user complains that "this contract procedure is very cumbersome," the emotion engine recognizes this dissatisfaction. The server then attempts to alleviate the user's dissatisfaction by providing simplified procedural information.

[0588] An example of a prompt to input into a generative AI model is, "Please tell me an effective way to respond when a user expresses dissatisfaction." Based on this prompt, the system will suggest an appropriate solution.

[0589] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0590] Step 1:

[0591] The server retrieves past business history from the information storage medium. Specifically, the server accesses the database using SQL queries and selects contract information and response history. The input is the business history in the database, and the output is the raw data imported into the server.

[0592] Step 2:

[0593] The server converts the acquired business history data into an information structure. Specifically, it performs actions such as converting the data format, deleting unnecessary fields, and filling in missing data. The input is the raw data acquired in step 1, and the output is the cleansed information structure.

[0594] Step 3:

[0595] The server supplies the cleansed information structure to the generating AI model and trains the model. Specifically, it formats the data appropriately and trains the model using machine learning libraries. The input is the information structure, and the output is the trained machine learning model.

[0596] Step 4:

[0597] The user sends an inquiry using a device. Specifically, the user inputs the inquiry via text or voice and presses the send button on the device. The input is the user's inquiry, and the output is the inquiry data sent from the device to the server.

[0598] Step 5:

[0599] The terminal sends inquiry data to the emotion engine. Specifically, the terminal performs a function that transfers the inquiry content to an analysis engine. The input is the user's inquiry data, and the output is the analysis result from the emotion engine.

[0600] Step 6:

[0601] The emotion engine analyzes user inquiries and recognizes emotional information. Specifically, it uses natural language processing to generate emotion vectors from input data. The input is user inquiry data, and the output is identified emotional information.

[0602] Step 7:

[0603] The server uses recognized sentiment information to leverage a machine learning model to generate the optimal response. Specifically, the server generates response candidates from the model and selects the one that best matches the sentiment. The input is sentiment information and query data, and the output is the optimized response.

[0604] Step 8:

[0605] The server sends the generated, optimized response to the terminal. Specifically, the server transmits the response to the terminal via network communication. The input is the response data generated within the server, and the output is the response displayed on the terminal.

[0606] Step 9:

[0607] Users review the answers presented on their devices and provide feedback. Specifically, users input satisfaction ratings and comments through a feedback form. The input is feedback data sent to the server, and the output is the collected user feedback.

[0608] Step 10:

[0609] The server improves the system based on the collected feedback. Specifically, it analyzes the feedback data and adjusts the emotion engine and machine learning models as needed. The input is user feedback data, and the output is improved model accuracy and corresponding techniques.

[0610] (Application Example 2)

[0611] Next, we will explain application example 2. In the following explanation, 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."

[0612] In systems used within the home, a challenge is to appropriately understand user emotions and provide more natural and effective interactions. Current systems do not take user emotions into consideration, which can lead to a decrease in the quality of the user experience.

[0613] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0614] In this invention, the server includes means for obtaining past behavioral history from an information set, means for converting the obtained behavioral history into a data structure, and means for supplying the data structure to a learning algorithm and training the algorithm. This makes it possible to analyze the user's emotions and generate an optimal response that takes the analysis results into account.

[0615] "Past behavioral history" refers to records and data about the user's past actions and interactions.

[0616] An "information collection" is a collection of diverse information and data stored in a database or storage device.

[0617] A "data structure" is a way of organizing and arranging data in a specific format so that it can be processed efficiently by a computer.

[0618] A "learning algorithm" is a series of computational procedures that enable a machine to autonomously learn patterns and knowledge from data.

[0619] An "information request" is an action initiated by a user to ask questions or seek support.

[0620] An "output device" is a device or interface used to present information generated by a system to the user.

[0621] To realize this application, the system is constructed as follows: The server retrieves past behavioral history from an information set and converts it into a data structure. Next, this data structure is supplied to a learning algorithm to train the algorithm. As the learning algorithm, general machine learning frameworks and tools can be used. Specific software that can be used include TensorFlow and PyTorch.

[0622] The terminal receives information requests from the user and sends them to the server. The server analyzes the received information requests using a learning algorithm and generates the optimal response. In this process, an emotion recognition engine is also used to analyze the user's emotions. The emotion recognition engine detects the user's emotions from voice input and text input, and uses the analysis results to generate the optimal response.

[0623] The generated response is sent back to the terminal and presented to the user via an output device. This output device may be a robot's display or speaker. The user reacts to the provided response, and this feedback is collected again by the server. The server uses this feedback to improve its learning algorithm and emotion recognition engine, contributing to improved response accuracy in subsequent interactions.

[0624] For example, if a user requests information saying "I'm tired today," the server receives this and uses its emotion recognition engine to determine that the user is feeling tired. Based on this, it generates suggestions for relaxing music to play and presents them to the user through the device.

[0625] An example of a prompt message would be, "Please tell me how to identify emotions from user input and suggest appropriate relaxation methods if the user is fatigued."

[0626] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0627] Step 1:

[0628] The server retrieves past behavioral history from a set of information. Input data includes logs of past user interactions. The server organizes this data item by item and retrieves it from storage in a specific format. The output is a structured dataset of behavioral history.

[0629] Step 2:

[0630] The server converts the acquired behavioral history into a data structure. The input data is the behavioral history dataset created in step 1. The server converts this into a data frame or other data structure and performs denoising and cleaning processes. The output is a clean dataset suitable for the learning algorithm.

[0631] Step 3:

[0632] The server feeds the clean dataset obtained in the previous step to the learning algorithm and trains the algorithm. The input data is the dataset that was the output of step 2. The server uses TensorFlow or PyTorch to tune the parameters of the learning algorithm and build the model. The output is a trained model suitable for a specific task.

[0633] Step 4:

[0634] The terminal receives information requests from the user. Input data is direct input from the user in voice or text format. The terminal prepares this data to be sent to the server as text. Output is the information request data ready for transmission.

[0635] Step 5:

[0636] The server analyzes the received information request using a learning algorithm and generates the optimal response. The input data is the information request sent in step 4. The server analyzes the user's emotions using an emotion recognition engine and calculates the optimal response based on the analysis results. The output is the response data to be presented to the user.

[0637] Step 6:

[0638] The terminal presents the generated response to the user through an output device. The input data is the response data generated in step 5. The terminal performs operations to express the response to the user using a speaker or display. The output is the response in a format that the user can see or hear.

[0639] Step 7:

[0640] The user responds to the presented response. The input is the user's response and feedback. The terminal acquires this input and prepares it for analysis by sending it to the server. The output is the feedback data sent to the server.

[0641] Step 8:

[0642] The server uses feedback to improve the learning algorithm and emotion recognition engine. The input data is the feedback data acquired in step 7. The server analyzes the feedback and improves the system's response accuracy by retraining the learning model and adjusting its parameters. The output is the improved learning model and emotion recognition engine.

[0643] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0644] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0645] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0646] [Fourth Embodiment]

[0647] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0648] As shown in Figure 7, the 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.

[0649] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0650] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0651] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0652] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0653] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0654] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0655] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0656] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0657] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0658] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0659] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0660] This invention is a system for streamlining and standardizing base station contract operations, utilizing past operational history and enabling the automatic generation of optimal responses to inquiries using machine learning models. Its specific operation is described below.

[0661] The server connects to the database and retrieves past contracts and correspondence history. This data is then formatted by the server into a format suitable for machine learning. The formatted data is then fed by the server to a machine learning model for training. The trained model learns patterns specific to contract work and has the ability to respond quickly to complex inquiries.

[0662] When a user enters a contract-related inquiry using their device, the device sends the inquiry to the server. The server passes the received inquiry to a machine learning model for analysis. As a result of this analysis, an answer optimized for contract operations is generated, and the server sends that answer to the device.

[0663] Users can view the answers provided on their devices. If the answers are unsatisfactory, users can provide feedback. This feedback is collected by the server and used to further improve the model.

[0664] As a concrete example, consider a case where a user asks, "Please tell me the appropriate contract process for installing a new base station." In this case, the server uses a model based on similar past cases to generate a response that includes the optimal process steps, and delivers it to the terminal. The user can then quickly proceed with the contract procedures based on these steps. Through this entire process, the system prevents contract work from becoming dependent on individual employees and enables consistent responses across the entire organization.

[0665] The following describes the processing flow.

[0666] Step 1:

[0667] The server connects to the database and retrieves past contract templates and correspondence history. This prepares the server to gather the necessary data.

[0668] Step 2:

[0669] The server converts the acquired data into a format suitable for machine learning. By removing noise and cleaning the data, a high-quality dataset is constructed.

[0670] Step 3:

[0671] The server supplies data to a machine learning model and trains the model. The server trains the model to learn patterns in base station contract operations.

[0672] Step 4:

[0673] The user enters their contract-related inquiry using a terminal. The user enters their specific question or request into the terminal and presses the send button.

[0674] Step 5:

[0675] The terminal sends the inquiry details to the server. The terminal formats the user's data appropriately and transfers it.

[0676] Step 6:

[0677] The server passes the received query to a machine learning model for analysis. The model generates the optimal response based on the trained data.

[0678] Step 7:

[0679] The server sends the generated response to the terminal. The server then sends the response back to the terminal in an appropriate format so that the user can easily understand it.

[0680] Step 8:

[0681] The terminal displays the answer to the user. This allows the user to confirm the answer to their inquiry and use it to their advantage in their work.

[0682] Step 9:

[0683] The user enters feedback on the answer into their device. This feedback may include expressions of satisfaction, suggestions for improvement, etc.

[0684] Step 10:

[0685] The device sends feedback to the server. The device formats the feedback correctly and passes it to the server.

[0686] Step 11:

[0687] The server collects feedback and uses it to improve the machine learning model. This improves the overall accuracy of the system, making future inquiries more accurate and faster.

[0688] (Example 1)

[0689] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0690] In traditional contract management, handling complex inquiries is often dependent on individual employees, resulting in insufficient efficiency and standardization of operations. Therefore, while consistency of information and prompt responses are required, obtaining uniform and optimal answers remains a challenge.

[0691] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0692] In this invention, the server includes means for acquiring past data history from an information storage device, means for converting the acquired data history into an information format, and means for analyzing received inquiries using a predictive analytics model and generating responses. This enables the streamlining and standardization of contract operations and provides quick and consistent answers even to complex inquiries.

[0693] "Past data history" refers to information such as contracts and correspondence records generated in the past during contract operations.

[0694] An "information storage device" is a system that organizes and stores data, and is used by users to retrieve information later.

[0695] "Information format" refers to data that has been transformed into a specific structure or format to be adapted for machine learning and analysis.

[0696] A "predictive analytics model" refers to an algorithm that uses machine learning techniques to learn patterns from past data and predict future inquiries and situations.

[0697] An "inquiry" refers to a question or request that a user enters into the system, seeking specific information or guidance on procedures.

[0698] A "response" is the result generated by a predictive analytics model based on its analysis, and includes specific information and instructions provided in response to a user's inquiry.

[0699] "Opinions" refer to feedback and evaluations of responses provided by users, and are used to improve the system.

[0700] "Denoising" refers to the process of removing unnecessary elements and inaccurate information from data, transforming it into a format suitable for analysis.

[0701] "Content organization" refers to processing and adjustments made to improve the consistency and clarity of data.

[0702] This invention is an advanced information processing system for streamlining and standardizing contract management. This enables rapid and accurate responses and prevents reliance on individual expertise in the work. The system primarily consists of the following components:

[0703] The server connects to an information storage device to retrieve historical data. A database management system is used for this, with PostgreSQL and MySQL being typical examples. After retrieving the data, the server converts it into an information format using the pandas library. At this stage, processing is performed to remove noise and organize the content.

[0704] Next, the server feeds the formatted data into a predictive analytics model, which is then trained using scikit-learn or TensorFlow. This model learns patterns specific to contract work and is then adapted to handle a variety of inquiries.

[0705] The user enters specific inquiries into the system via their terminal. For example, a prompt might read, "Please tell me the appropriate contract process for installing a new base station." This inquiry is then sent from the terminal to the server.

[0706] Subsequently, the server analyzes the query using a predictive analytics model and generates the optimal response. This response is based on similar past cases and provides consistent information to the user. The generated response is sent from the server to the terminal, where the user can review it.

[0707] If the response provided is unsatisfactory, users can provide feedback. This feedback is aggregated on the server and used to improve the model's performance. This allows the system to continuously improve, increasing the efficiency and consistency of contract management.

[0708] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0709] Step 1:

[0710] The server connects to the information storage device and retrieves historical data. The input is a contract-related dataset, and the data is extracted by executing SQL queries using a database management system. The output is a set of raw data. The goal of this step is to efficiently and accurately collect the necessary data.

[0711] Step 2:

[0712] This step converts the raw data acquired by the server into an informational format. The input is the raw data obtained in step 1, which is converted into a data frame using the pandas library, and then denoised and organized. The output is clean and structured data. Specific actions in this step include removing whitespace values ​​and normalizing text.

[0713] Step 3:

[0714] The server trains a predictive analytics model using the formatted data. The input is the formatted data from step 2, and the model is trained using libraries such as scikit-learn and TensorFlow. The output is the trained model. In this process, the model learns the patterns necessary for contract work.

[0715] Step 4:

[0716] The user enters specific prompt text on their terminal and sends an inquiry about the contract. The input is a prompt text, for example, "Please tell me the appropriate contract process for installing a new base station." The output is a structured data request sent to the server.

[0717] Step 5:

[0718] The server receives the prompt and performs analysis using a predictive analytics model. The input is the structured data request from step 4, and the model performs analysis using reference data. The output is an optimized response to the query. This analysis involves inference based on similar past cases.

[0719] Step 6:

[0720] The server sends the generated response to the user's terminal. The input is the response data generated in step 5, and the output is the information displayed on the user's terminal. Appropriate presentation of the response content is important here.

[0721] Step 7:

[0722] The user provides feedback on the response they receive and sends it from their terminal to the server. The input is the user's feedback data, and the output is aggregated information of the opinions on the server. This feedback contributes to improving the model and making decisions for the next steps.

[0723] (Application Example 1)

[0724] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0725] In traditional contract management, past business history could not be efficiently utilized, resulting in inefficient and person-dependent responses to various inquiries. Furthermore, even when there was a demand for improved response quality and faster response times, there was a lack of a system to properly manage and analyze the necessary data. As a result, providing immediate voice responses to user inquiries was difficult, and the system lacked flexibility and responsiveness.

[0726] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0727] In this invention, the server includes means for acquiring past history from an information management device, means for converting the acquired history into an information structure, means for supplying the information structure to a learning device and training the device, means for collecting voice input and converting voice into text information, and means for providing the generated response to the user as voice output. This makes it possible to respond to user inquiries quickly and accurately, thereby achieving efficiency and standardization of contract operations.

[0728] "Past history" refers to records of a series of operations and transactions that have taken place in the past, and is the information stored in the data management device.

[0729] An "information management device" is a device used for acquiring, storing, retrieving, and processing data, and includes database systems.

[0730] "Information structure" refers to the form of data that has been formatted in a way that can be handled by machine learning devices, and is a data format suitable for analysis and training.

[0731] A "learning device" is a computer system that uses machine learning algorithms to process data and perform analysis and predictions.

[0732] The term "user" refers to the entity that uses a system or service to perform contractual work or related tasks, and is usually a human operator or person in charge.

[0733] "Voice input" is an input method that electronically collects the voice spoken by the user and converts it into analyzable data.

[0734] "Converting audio to text information" is the process of converting collected audio data into text data using natural language processing technology.

[0735] "Voice output" refers to an output method for transmitting generated responses or information to the user in voice format, and speech synthesis technology supports this.

[0736] When implementing this invention, a high-performance server and a terminal such as a smartphone or smart glasses are required. In this embodiment, the system operates in the following steps.

[0737] The server first retrieves historical data from the information management device. This data is transformed into an information structure, and the data is formatted to minimize noise. This information structure is then supplied to a training device using a machine learning framework such as TensorFlow, and the device is trained. Through this training, the device acquires advanced analytical capabilities for queries.

[0738] Users make inquiries using voice input via smartphones or smart glasses. This voice is converted into text information within the device and sent to the server. The server uses a trained learning device to analyze the inquiry and generate the optimal response.

[0739] The generated response is sent back from the server to the user's terminal and provided to the user as voice output using speech synthesis technology. This entire process allows the user to quickly obtain information about contract-related questions and transaction details.

[0740] For example, when a user asks "Please let me know the payment due date for next month" through smart glasses, the system can calculate the payment timing from the history and inform the user via voice, "The next payment due date is May 15th."

[0741] An example of a prompt to a generative AI model would be, "Based on my recent payment history, please provide a detailed, step-by-step process for my next payment." By using this prompt, the AI ​​model can use past history to derive an accurate answer.

[0742] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0743] Step 1:

[0744] The server retrieves historical data from the information management device. At this stage, the input is past business data stored within the information management device, and all relevant information is passed to the server as output. The server analyzes the retrieved data and formats it into the format required for subsequent processing.

[0745] Step 2:

[0746] The server uses the formatted historical data to transform it into an information structure. The input here is the historical data obtained in step 1, and the output is a dataset that can be processed by a machine learning framework. Through data format transformation and refinement, clean data with noise removed is generated.

[0747] Step 3:

[0748] The server uses TensorFlow to supply information structures to the learning machine and train the learning model. The input in this step is the clean dataset prepared in step 2, and the output is a trained learning model suitable for query analysis. The model is optimized using advanced algorithms, leveraging the computing resources on the server side.

[0749] Step 4:

[0750] The user uses a smartphone or smart glasses to perform voice input. This input consists of the user's inquiry. The device converts the voice into text and sends it to the server. The output is the user's inquiry converted into text.

[0751] Step 5:

[0752] The server analyzes the received character information using a learning device and generates the optimal response. In this process, the input is the character information obtained in step 4, and the output is the generated response. Considering the prompt text and past history, the AI ​​model provides an accurate answer to the query.

[0753] Step 6:

[0754] The server sends the generated response to the terminal. The input is the response generated in step 5, and the output is the voice output delivered to the user using speech synthesis technology. The terminal uses the received response to provide real-time voice feedback to the user.

[0755] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0756] This invention relates to a system that not only improves the efficiency of base station contract operations but also enables more natural and optimal responses by recognizing user emotions. This system incorporates an emotion engine that utilizes past business history and enables responses that take user emotions into consideration.

[0757] The server retrieves past contract information and response history from the database, cleanses this data, and provides it to a machine learning model. Through this process, the model learns knowledge about contract operations and becomes able to generate highly accurate responses to different inquiries.

[0758] When a user submits an inquiry using a device, the device first sends the information to the emotion engine. The emotion engine analyzes the user's input text and voice to recognize their emotions. The server combines this emotion information to generate the most appropriate response based on the inquiry.

[0759] The generated response is sent from the server to the terminal and presented to the user. User feedback is also collected and used to improve both the sentiment engine and the machine learning model, further enhancing future responses.

[0760] As a concrete example, consider a scenario where a user expresses dissatisfaction, stating, "This contract procedure is extremely cumbersome." The emotion engine recognizes the user's dissatisfaction and determines that an urgent solution is needed. Based on this, the server presents simplified procedures or common solutions. In this way, utilizing an emotion engine makes it possible to improve the user experience.

[0761] The following describes the processing flow.

[0762] Step 1:

[0763] The server connects to the database and retrieves past contract information and service history. The server collects this data and prepares for the next processing step.

[0764] Step 2:

[0765] The server denoises and cleanses the acquired data, converting it into a format suitable for machine learning models. The server then stores this cleansed data.

[0766] Step 3:

[0767] The server supplies data to a machine learning model and trains the model. The server enables the model to learn contract patterns and generate accurate answers to inquiries.

[0768] Step 4:

[0769] The user enters and submits a contract-related inquiry using their device. The user clearly enters the necessary information and questions.

[0770] Step 5:

[0771] The device sends inquiry information to the emotion engine, which analyzes the user's emotional state. The emotion engine infers emotions from the user's text and voice.

[0772] Step 6:

[0773] The server receives emotion data and query content from the emotion engine, inputs the received information into a machine learning model for analysis, and generates a response adapted to the user's emotions.

[0774] Step 7:

[0775] The server sends the generated response to the terminal. The user receives the response through the terminal and confirms its contents.

[0776] Step 8:

[0777] Users enter feedback on their devices. This feedback includes their satisfaction with the response and suggestions for improvement.

[0778] Step 9:

[0779] The device sends feedback to the server. The server receives the feedback and uses it to improve the sentiment engine and machine learning models.

[0780] Step 10:

[0781] The server analyzes the feedback and adjusts the machine learning model and sentiment engine. This improves the accuracy and emotional response of the next user response.

[0782] (Example 2)

[0783] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0784] In traditional contract management, handling user inquiries was inefficient, and it was difficult to provide optimal responses while considering user emotions. Furthermore, there was insufficient mechanism for effectively utilizing user feedback to improve the system. As a result, the user experience was compromised, and operational productivity suffered.

[0785] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0786] In this invention, the server includes means for acquiring past business history from an information storage medium, means for converting the acquired business history into an information structure, means for supplying the information structure to a machine learning model and training the model, means for analyzing user input and recognizing emotional information, and means for generating an optimal response based on the emotional information. This makes it possible to generate natural and optimal responses that take the user's emotions into consideration, thereby improving the user experience.

[0787] An "information storage medium" is a medium from which data can be stored and retrieved as needed.

[0788] "Information structure" refers to the organization and arrangement of data into a format that is easily usable for a specific purpose.

[0789] A "machine learning model" is a computational model that learns from data and performs pattern recognition and classification.

[0790] "Emotional information" refers to information extracted from user input data to identify the state and type of emotions.

[0791] An "optimal response" is one that generates the most appropriate and effective response based on the user's inquiry and emotional information.

[0792] This invention is a system that improves the efficiency of contract work and the user experience by taking user emotions into consideration and providing optimal responses. Implementation primarily involves servers, terminals, and users.

[0793] The server retrieves past business history from the data storage medium. This data is converted into an information structure using data analysis software. This information structure is then fed into a generative AI model, which is trained to handle various queries. The software used includes data cleansing tools and machine learning libraries.

[0794] Users submit inquiries via text or voice using their devices. These inquiries are sent by the device to the sentiment engine. The sentiment engine uses natural language processing software to analyze the user's emotions. The emotion information is sent to a server, which uses machine learning models to generate the most appropriate response based on this information.

[0795] The generated answers are presented to the user via the device. User feedback is sent to the server and used to improve the entire system.

[0796] For example, if a user complains that "this contract procedure is very cumbersome," the emotion engine recognizes this dissatisfaction. The server then attempts to alleviate the user's dissatisfaction by providing simplified procedural information.

[0797] An example of a prompt to input into a generative AI model is, "Please tell me an effective way to respond when a user expresses dissatisfaction." Based on this prompt, the system will suggest an appropriate solution.

[0798] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0799] Step 1:

[0800] The server retrieves past business history from the information storage medium. Specifically, the server accesses the database using SQL queries and selects contract information and response history. The input is the business history in the database, and the output is the raw data imported into the server.

[0801] Step 2:

[0802] The server converts the acquired business history data into an information structure. Specifically, it performs actions such as converting the data format, deleting unnecessary fields, and filling in missing data. The input is the raw data acquired in step 1, and the output is the cleansed information structure.

[0803] Step 3:

[0804] The server supplies the cleansed information structure to the generating AI model and trains the model. Specifically, it formats the data appropriately and trains the model using machine learning libraries. The input is the information structure, and the output is the trained machine learning model.

[0805] Step 4:

[0806] The user sends an inquiry using a device. Specifically, the user inputs the inquiry via text or voice and presses the send button on the device. The input is the user's inquiry, and the output is the inquiry data sent from the device to the server.

[0807] Step 5:

[0808] The terminal sends inquiry data to the emotion engine. Specifically, the terminal performs a function that transfers the inquiry content to an analysis engine. The input is the user's inquiry data, and the output is the analysis result from the emotion engine.

[0809] Step 6:

[0810] The emotion engine analyzes user inquiries and recognizes emotional information. Specifically, it uses natural language processing to generate emotion vectors from input data. The input is user inquiry data, and the output is identified emotional information.

[0811] Step 7:

[0812] The server uses recognized sentiment information to leverage a machine learning model to generate the optimal response. Specifically, the server generates response candidates from the model and selects the one that best matches the sentiment. The input is sentiment information and query data, and the output is the optimized response.

[0813] Step 8:

[0814] The server sends the generated, optimized response to the terminal. Specifically, the server transmits the response to the terminal via network communication. The input is the response data generated within the server, and the output is the response displayed on the terminal.

[0815] Step 9:

[0816] Users review the answers presented on their devices and provide feedback. Specifically, users input satisfaction ratings and comments through a feedback form. The input is feedback data sent to the server, and the output is the collected user feedback.

[0817] Step 10:

[0818] The server improves the system based on the collected feedback. Specifically, it analyzes the feedback data and adjusts the emotion engine and machine learning models as needed. The input is user feedback data, and the output is improved model accuracy and corresponding techniques.

[0819] (Application Example 2)

[0820] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0821] In systems used within the home, a challenge is to appropriately understand user emotions and provide more natural and effective interactions. Current systems do not take user emotions into consideration, which can lead to a decrease in the quality of the user experience.

[0822] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0823] In this invention, the server includes means for obtaining past behavioral history from an information set, means for converting the obtained behavioral history into a data structure, and means for supplying the data structure to a learning algorithm and training the algorithm. This makes it possible to analyze the user's emotions and generate an optimal response that takes the analysis results into account.

[0824] "Past behavioral history" refers to records and data about the user's past actions and interactions.

[0825] An "information collection" is a collection of diverse information and data stored in a database or storage device.

[0826] A "data structure" is a way of organizing and arranging data in a specific format so that it can be processed efficiently by a computer.

[0827] A "learning algorithm" is a series of computational procedures that enable a machine to autonomously learn patterns and knowledge from data.

[0828] An "information request" is an action initiated by a user to ask questions or seek support.

[0829] An "output device" is a device or interface used to present information generated by a system to the user.

[0830] To realize this application, the system is constructed as follows: The server retrieves past behavioral history from an information set and converts it into a data structure. Next, this data structure is supplied to a learning algorithm to train the algorithm. As the learning algorithm, general machine learning frameworks and tools can be used. Specific software that can be used include TensorFlow and PyTorch.

[0831] The terminal receives information requests from the user and sends them to the server. The server analyzes the received information requests using a learning algorithm and generates the optimal response. In this process, an emotion recognition engine is also used to analyze the user's emotions. The emotion recognition engine detects the user's emotions from voice input and text input, and uses the analysis results to generate the optimal response.

[0832] The generated response is sent back to the terminal and presented to the user via an output device. This output device may be a robot's display or speaker. The user reacts to the provided response, and this feedback is collected again by the server. The server uses this feedback to improve its learning algorithm and emotion recognition engine, contributing to improved response accuracy in subsequent interactions.

[0833] For example, if a user requests information saying "I'm tired today," the server receives this and uses its emotion recognition engine to determine that the user is feeling tired. Based on this, it generates suggestions for relaxing music to play and presents them to the user through the device.

[0834] An example of a prompt message would be, "Please tell me how to identify emotions from user input and suggest appropriate relaxation methods if the user is fatigued."

[0835] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0836] Step 1:

[0837] The server retrieves past behavioral history from a set of information. Input data includes logs of past user interactions. The server organizes this data item by item and retrieves it from storage in a specific format. The output is a structured dataset of behavioral history.

[0838] Step 2:

[0839] The server converts the acquired behavioral history into a data structure. The input data is the behavioral history dataset created in step 1. The server converts this into a data frame or other data structure and performs denoising and cleaning processes. The output is a clean dataset suitable for the learning algorithm.

[0840] Step 3:

[0841] The server feeds the clean dataset obtained in the previous step to the learning algorithm and trains the algorithm. The input data is the dataset that was the output of step 2. The server uses TensorFlow or PyTorch to tune the parameters of the learning algorithm and build the model. The output is a trained model suitable for a specific task.

[0842] Step 4:

[0843] The terminal receives information requests from the user. Input data is direct input from the user in voice or text format. The terminal prepares this data to be sent to the server as text. Output is the information request data ready for transmission.

[0844] Step 5:

[0845] The server analyzes the received information request using a learning algorithm and generates the optimal response. The input data is the information request sent in step 4. The server analyzes the user's emotions using an emotion recognition engine and calculates the optimal response based on the analysis results. The output is the response data to be presented to the user.

[0846] Step 6:

[0847] The terminal presents the generated response to the user through an output device. The input data is the response data generated in step 5. The terminal performs operations to express the response to the user using a speaker or display. The output is the response in a format that the user can see or hear.

[0848] Step 7:

[0849] The user responds to the presented response. The input is the user's response and feedback. The terminal acquires this input and prepares it for analysis by sending it to the server. The output is the feedback data sent to the server.

[0850] Step 8:

[0851] The server uses feedback to improve the learning algorithm and emotion recognition engine. The input data is the feedback data acquired in step 7. The server analyzes the feedback and improves the system's response accuracy by retraining the learning model and adjusting its parameters. The output is the improved learning model and emotion recognition engine.

[0852] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0853] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0854] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0855] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0856] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0857] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0858] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0859] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0860] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0861] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0862] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0863] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0864] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0865] 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.

[0866] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0867] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0868] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0869] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0870] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0871] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0872] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0873] The following is further disclosed regarding the embodiments described above.

[0874] (Claim 1)

[0875] Methods for retrieving past work history from a database,

[0876] A means of converting acquired business history into a data structure,

[0877] A means for supplying the aforementioned data structure to a machine learning model and training the model,

[0878] Means of receiving inquiries from users,

[0879] A means for analyzing an inquiry received using the aforementioned machine learning model and generating an answer,

[0880] A means of sending the generated response to the user's terminal,

[0881] A means for aggregating user feedback on received responses and using it to improve the machine learning model,

[0882] A system that includes this.

[0883] (Claim 2)

[0884] The system according to claim 1, further comprising means for denoising and cleaning the data structure.

[0885] (Claim 3)

[0886] The system according to claim 1, further comprising means for adjusting the machine learning model so that it can respond to multiple different queries.

[0887] "Example 1"

[0888] (Claim 1)

[0889] A means of obtaining past data history from an information storage device,

[0890] A means of converting acquired data history into an information format,

[0891] A means for supplying the aforementioned information format to a predictive analysis model and training the model,

[0892] Means of receiving inquiries from users,

[0893] A means for analyzing an inquiry received using the aforementioned predictive analysis model and generating a response,

[0894] A means for sending the generated response to the user's terminal,

[0895] A means for aggregating user feedback on received responses and using it to improve the predictive analytics model,

[0896] A means of generating the optimal response based on similar past cases,

[0897] A system that includes this.

[0898] (Claim 2)

[0899] The system according to claim 1, further comprising means for denoising and organizing the content of the aforementioned information format.

[0900] (Claim 3)

[0901] The system according to claim 1, further comprising means for adjusting the predictive analytics model so that it can respond to a variety of inquiries.

[0902] "Application Example 1"

[0903] (Claim 1)

[0904] A means of obtaining past history from an information management device,

[0905] A means of converting acquired history into an information structure,

[0906] Means for supplying the aforementioned information structure to a learning device and training the device,

[0907] A means of receiving inquiries from users,

[0908] A means for analyzing an inquiry received using the aforementioned learning device and generating a response,

[0909] Means for transmitting the generated response to the user device,

[0910] A means for collecting user evaluations of received responses and using them to improve the learning device,

[0911] A means for collecting voice input and converting it into text information,

[0912] A means of providing the generated response to the user via audio output,

[0913] A system that includes this.

[0914] (Claim 2)

[0915] The system according to claim 1, further comprising means for removing unnecessary information and purifying information in the aforementioned information structure.

[0916] (Claim 3)

[0917] The system according to claim 1, further comprising means for adjusting the learning device so that it can respond to a plurality of different queries.

[0918] "Example 2 of combining an emotion engine"

[0919] (Claim 1)

[0920] Methods for obtaining past business history from information storage media,

[0921] A means of converting acquired business history into an information structure,

[0922] A means for supplying the aforementioned information structure to a machine learning model and training the model,

[0923] Means of receiving inquiries from users,

[0924] A means for analyzing an inquiry received using the aforementioned machine learning model and generating an answer,

[0925] A means of sending the generated response to the user's terminal,

[0926] A means for aggregating user feedback on received responses and using it to improve the machine learning model,

[0927] A means of analyzing user input to recognize emotional information,

[0928] A means for generating an optimal response based on the aforementioned emotional information,

[0929] A system that includes this.

[0930] (Claim 2)

[0931] The system according to claim 1, further comprising means for performing noise reduction and cleaning on the information structure.

[0932] (Claim 3)

[0933] The system according to claim 1, further comprising means for adjusting the machine learning model so that it can respond to multiple different queries.

[0934] "Application example 2 when combining with an emotional engine"

[0935] (Claim 1)

[0936] A means of obtaining past behavioral history from an information set,

[0937] A means of converting acquired behavioral history into a data structure,

[0938] A means for supplying the aforementioned data structure to a learning algorithm and training the algorithm,

[0939] Means of receiving information requests from users,

[0940] A means for analyzing an information request received using the aforementioned learning algorithm and generating a response,

[0941] Means for transmitting the generated response to an output device,

[0942] A means for aggregating the user's responses to received responses and using them to improve the learning algorithm,

[0943] A means for analyzing the user's emotions and optimizing the response by considering the analysis results,

[0944] A system that includes this.

[0945] (Claim 2)

[0946] The system according to claim 1, further comprising means for removing unwanted components and purifying the data structure.

[0947] (Claim 3)

[0948] The system according to claim 1, further comprising means for adjusting the learning algorithm so that it can respond to multiple different information requests. [Explanation of symbols]

[0949] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Methods for retrieving past work history from a database, A means of converting acquired business history into a data structure, A means for supplying the aforementioned data structure to a machine learning model and training the model, Means of receiving inquiries from users, A means for analyzing an inquiry received using the aforementioned machine learning model and generating an answer, A means of sending the generated response to the user's terminal, A means for aggregating user feedback on received responses and using it to improve the machine learning model, A system that includes this.

2. The system according to claim 1, further comprising means for denoising and cleaning the data structure.

3. The system according to claim 1, further comprising means for adjusting the machine learning model so that it can respond to multiple different queries.

Images