System

Abstract

To provide a system.SOLUTION: A system includes: means for searching for document data related to input information of a user using a natural language processing engine; means for analyzing an emotional context of the user using a feeling engine; means for sorting search results according to the emotional context of the user; and means for providing the user with the sorted search results.SELECTED DRAWING: Figure 1

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, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of the chatbot's character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] For employees to efficiently search for and organize the necessary information is an important issue in the modern business environment where there are a large number of documents and materials. However, in the conventional search systems, it has been difficult for employees to accurately search for and organize the necessary information.

Means for Solving the Problems

[0005] The present invention uses a natural language processing engine to search for relevant documents and materials from the questions and keywords of employees and provide the search results to the employees. Furthermore, the documents and materials are organized based on the search results. Thereby, it becomes possible for employees to efficiently search for and organize the necessary information.

Brief Description of the Drawings

[0006]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Embodiments for Carrying Out the Invention

[0007] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0008] First, the language used in the following description will be explained.

[0009] In the following embodiments, the labeled 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), or a TPU (TENSOR PROCESSING UNIT (registered trademark)), etc.

[0010] In the following embodiments, the labeled RAM (Random Access Memory) is a memory where information is temporarily stored and is used as a work memory by the processor.

[0011] In the following embodiments, the labeled 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, etc.

[0012] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), etc.

[0013] 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 only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0014] [First Embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] As shown in FIG. 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.

[0017] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of the "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. Also, the database 24 and the communication I / F 26 are connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0018] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. Also, the reception device 38, the output device 40, and the camera 42 are connected to the bus 52.

[0019] The reception device 38 includes a touch panel 38A, a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by contact of an indicator (for example, a pen or a finger, etc.) by detecting the contact of the indicator. The microphone 38B receives user input by voice by detecting the voice of the user. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires data indicating the user input.

[0020] The output device 40 includes a display 40A, a speaker 40B, etc., and presents data to the user 20 by outputting the data in a form (for example, voice and / or text) that can be perceived by the user 20. The display 40A displays visible information such as text and images according to an instruction from the processor 46. The speaker 40B outputs voice according to an instruction from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, a diaphragm, and a shutter, and an imaging device such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0021] The communication I / F 44 is connected to the network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] As shown in FIG. 2, in the data processing apparatus 12, specific processing is performed by the processor 28. The storage 32 stores a specific processing program 56. The specific processing program 56 is an example of the "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0024] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the specific processing unit 290.

[0025] In the smart device 14, reception / output processing is performed by the processor 46. The storage 50 stores a reception / output program 60. The reception / output program 60 is used in combination with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception / output program 60 from the storage 50 and executes the read reception / output program 60 on the RAM 48. The reception / output 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing apparatus 12 will be described.

[0027] "Form Example 1"

[0028] As an embodiment of the present invention, the natural language processing engine analyzes an employee's questions and keywords based on a machine learning model. Specifically, when an employee inputs a keyword such as "latest sales report", the natural language processing engine analyzes this keyword and searches for related documents and materials from the database.

[0029] "Form Example 2"

[0030] The search results are provided to the employees. Specifically, the search results are displayed in a list format, and the employees can select the required documents and materials. Also, the search results are displayed in descending order of relevance, enabling the employees to quickly find the information they need.

[0031] "Form Example 3"

[0032] Furthermore, the documents and materials are organized based on the search results. Specifically, the search results are categorized and automatically stored in the corresponding folders or directories. For example, if there is a category of "Sales Report", the documents and materials related to this category are automatically stored in the "Sales Report" folder. This allows the employees to efficiently search for and organize the required information.

[0033] The processing flow of each form example is described below.

[0034] "Form Example 1"

[0035] Step 1: The employee inputs questions or keywords into the system. For example, keywords such as "latest sales report" are input.

[0036] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. This analysis is performed based on a machine learning model.

[0037] Step 3: The search results are displayed in a list format, and the employee can select the required documents and materials. Also, the search results are displayed in descending order of relevance.

[0038] "Form Example 2"

[0039] Step 1: The employee inputs questions or keywords into the system. For example, keywords such as "latest sales report" are input.

[0040] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. This analysis is performed based on a machine learning model.

[0041] Step 3: The search results are displayed in a list format, and employees can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance.

[0042] Step 4: The selected documents and materials are automatically stored in the folders or directories corresponding to their respective categories. For example, if there is a category of "sales report", the documents and materials related to this category are automatically stored in the "sales report" folder.

[0043] (Example 1)

[0044] Next, Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0045] In order for employees to perform their work efficiently, it is important to quickly search for and obtain the necessary documents and materials. However, in conventional systems, the accuracy of keyword searches was low, and it often took a long time to find relevant documents and materials. Also, since the sorting of search results was done manually, there were problems such as low efficiency and a high likelihood of errors. To solve these problems, a system with a more accurate search function and an automatic sorting function is needed.

[0046] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following respective means.

[0047] In this invention, the server includes means for using a natural language processing engine to search for relevant documents and materials from employees' questions and keywords, means for providing the search results to employees, means for sorting documents and materials based on the search results, means for employees to input keywords from a terminal, means for the terminal to send the input keywords to the server, means for the server to pass the keywords to the natural language processing engine, means for the natural language processing engine to analyze the keywords, means for the server to search a database based on the analysis results, means for the server to return the search results to the terminal, and means for the terminal to display the search results to employees. As a result, employees can quickly and accurately search for and obtain the necessary documents and materials. In addition, the automatic sorting function of the search results can improve work efficiency and reduce errors.

[0048] The "natural language processing engine" is software for analyzing human language using a machine learning model to understand its meaning.

[0049] An "employee" is an individual who belongs to a company or organization and conducts business.

[0050] A "question" is the content of an inquiry input by an employee to obtain information.

[0051] A "keyword" is a word or phrase input by an employee to search for specific information.

[0052] A "document" is a material such as a text file or report in which information related to business is described.

[0053] "Materials" are data and documents containing information related to business.

[0054] The "means for searching" is a method for identifying relevant documents and materials using a natural language processing engine.

[0055] The "means for providing" is a method for displaying search results to employees.

[0056] The "sorting means" is a method of categorizing documents and materials based on search results and storing them in appropriate folders or directories.

[0057] The "terminal" is a device such as a computer or smartphone used by employees.

[0058] The "server" is a computer system that receives requests from terminals and performs processing.

[0059] The "database" is an aggregate of information in which documents and materials are stored.

[0060] The "analysis means" is a method of understanding the meaning of keywords using a natural language processing engine and extracting relevant information.

[0061] The "search results" is a list of relevant documents and materials identified by a natural language processing engine.

[0062] The "display means" is a method of visually presenting search results on a terminal.

[0063] This invention is a system for quickly searching for and obtaining documents and materials necessary for employees to efficiently perform their work. This system has a function of analyzing employees' questions and keywords using a natural language processing engine and searching for relevant documents and materials from a database.

[0064] Hardware and software to be used

[0065] Hardware

[0066] Server: A high-performance computer system that manages a database and executes a natural language processing engine.

[0067] Terminal: A device such as a computer or smartphone used by employees.

[0068] Software

[0069] Natural language processing engine: Analyzes employees' questions and keywords using machine learning models (such as BERT, GPT-3 (registered trademark), etc.).

[0070] Database: An aggregate of information where documents and materials are stored, and for example, MySQL (registered trademark) etc. is used.

[0071] Data processing and data calculation

[0072] The server receives the keywords input by the employee from the terminal and sends them to the natural language processing engine. The natural language processing engine analyzes the keywords using a machine learning model and extracts information for identifying relevant documents and materials. The server searches the database based on the analysis results and retrieves relevant documents and materials. The retrieved search results are returned from the server to the terminal and displayed for the employee to view.

[0073] Specific example

[0074] As a specific example, consider the case where an employee inputs "latest sales report". In this case, the server performs the following processing.

[0075] 1. The user inputs "latest sales report" into the input field of the terminal.

[0076] 2. The terminal uses an HTTP POST request to send the keywords to the server.

[0077] 3. The server passes the keywords to the natural language processing engine.

[0078] 4. The natural language processing engine analyzes "latest sales report" using the BERT model.

[0079] 5. Based on the analysis result, the server searches the MySQL database to identify the latest business report.

[0080] 6. The server returns the search result to the terminal in JSON format.

[0081] 7. The terminal displays the search result to the user so that the user can view the latest business report.

[0082] Example of prompt sentence

[0083] The following are examples of prompt sentences entered by the user.

[0084] "Please display the latest business report."

[0085] "Please tell me the sales data for 2023."

[0086] "Please search for the marketing materials of the new product."

[0087] In this way, the user can easily obtain the required information.

[0088] The flow of the specific process in Example 1 will be described with reference to FIG. 11.

[0089] Step 1:

[0090] The user inputs a keyword from the terminal.

[0091] As a specific operation, the user inputs a keyword such as "latest business report" into the input field of the terminal. The input keyword is temporarily stored in the memory of the terminal.

[0092] Input: Keyword entered by the user (e.g., "latest business report")

[0093] Output: Keyword stored in the terminal

[0094] Step 2:

[0095] The terminal sends the input keyword to the server.

[0096] Specifically, the terminal uses an HTTP POST request to send the input keyword to the server. At this time, the keyword is included in the request body.

[0097] Input: Keyword stored in the terminal

[0098] Output: Keyword sent to the server

[0099] Step 3:

[0100] The server passes the keyword to the natural language processing engine.

[0101] Specifically, the server passes the received keyword to the natural language processing engine. At this time, the keyword is passed as a parameter of the API request.

[0102] Input: Keyword sent to the server

[0103] Output: Keyword passed to the natural language processing engine

[0104] Step 4:

[0105] The natural language processing engine analyzes the keyword.

[0106] Specifically, the natural language processing engine uses a machine learning model (e.g., BERT or GPT-3) to analyze the keyword. As an analysis result, information for identifying relevant documents and materials is generated.

[0107] Input: Keyword passed to the natural language processing engine

[0108] Output: Analysis result (information on relevant documents and materials)

[0109] Step 5:

[0110] The server searches the database based on the analysis results.

[0111] Specifically, the server searches a database (e.g., MySQL) based on the analysis results. A search query is generated based on the analysis results and executed against the database.

[0112] Input: Analysis results

[0113] Output: Search results (related documents and materials) retrieved from the database

[0114] Step 6:

[0115] The server returns the search results to the device.

[0116] Specifically, the server returns the search results obtained from the database to the terminal in JSON format, etc., as an HTTP response.

[0117] Input: Search results retrieved from the database

[0118] Output: Search results sent to your device

[0119] Step 7:

[0120] The terminal displays the search results to the user.

[0121] Specifically, the terminal displays the search results received from the server on the user interface, for example, displaying a link to the latest sales report and a preview of its contents.

[0122] Input: Search results sent to your device

[0123] Output: Search results (related documents and resources) displayed to the user

[0124] (Application Example 1)

[0125] Next, Application Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0126] In a logistics center, there is a problem that it is difficult for employees to quickly obtain necessary information. In particular, there is a lack of means to efficiently search for and provide information such as inventory status, delivery schedules, and the location of items in the warehouse. For this reason, work efficiency decreases, and there is a possibility of work delays and mistakes.

[0127] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0128] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for analyzing the questions input by employees in voice or text and searching for relevant information from a database, and means for displaying the search results on a smartphone. As a result, employees can quickly obtain the necessary information and improve work efficiency.

[0129] The "natural language processing engine" is software for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[0130] The "machine learning model" is an algorithm that learns based on data and analyzes questions and keywords.

[0131] The "database" is a collection of structured data for efficiently searching for and obtaining relevant information.

[0132] A "smartphone" is a portable electronic device that analyzes questions input in voice or text and displays search results.

[0133] "Search results" are relevant information retrieved from a database based on questions or keywords analyzed by a natural language processing engine.

[0134] "Documents and materials" are a collection of texts and data containing information required by employees.

[0135] "Categorization" is a process of classifying documents and materials based on search results and automatically storing them in folders or directories corresponding to each category.

[0136] "Questions input in voice or text" are the content of inquiries input by employees in voice or text format through a smartphone.

[0137] "Relevant information" is the necessary data and materials retrieved from a database for employees' questions and keywords.

[0138] The system for implementing this invention includes a natural language processing engine, a machine learning model, a database, and a smartphone. The specific configuration and operation of the system will be described below.

[0139] Configuration of the System

[0140] 1. Natural language processing engine: Software for analyzing questions and keywords input by employees and searching for relevant documents and materials. Specifically, natural language processing libraries such as spaCy are used.

[0141] 2. Machine learning model: An algorithm used to analyze questions and keywords. It is constructed using machine learning libraries such as scikit-learn.

[0142] 3. Database: A collection of structured data for efficiently searching and retrieving relevant information. Use a database management system such as SQLite.

[0143] 4. Smartphone: A portable electronic device used by employees to input questions in voice or text and display search results. An iOS or ANDROID (registered trademark) smartphone is used.

[0144] System Operation

[0145] 1. User Input: Employees input questions in voice or text format using a smartphone. For example, input a prompt sentence such as "Tell me the latest inventory status".

[0146] 2. Natural Language Processing: The server analyzes the user input using a natural language processing engine and extracts keywords. For example, keywords such as "inventory status" and "latest" are extracted.

[0147] 3. Database Search: The server searches the database based on the extracted keywords and retrieves relevant documents and materials. For example, a document containing the latest inventory information is searched.

[0148] 4. Provision of Search Results: The server sends the search results to the smartphone and displays them to the user. As a result, employees can quickly obtain the necessary information.

[0149] Specific Example

[0150] When an employee enters "Tell me the latest inventory status" into the smartphone, the natural language processing engine extracts keywords such as "inventory status" and "latest", and searches the database for the latest inventory information. As a result, the latest inventory information is displayed on the smartphone.

[0151] As examples of other prompt sentences, questions such as "Tell me the latest delivery schedule" or "Tell me the location of the items in the warehouse" can be considered.

[0152] With this system, employees can quickly obtain the necessary information and improve work efficiency.

[0153] The flow of the specific process in Application Example 1 will be described with reference to FIG. 12.

[0154] Step 1:

[0155] The user inputs a question in voice or text form using a smartphone.

[0156] Input: User's question in voice or text form (e.g., "Tell me the latest inventory status")

[0157] Output: Question data input into the smartphone

[0158] Specific operation: The user launches the smartphone application and performs voice input or text input. In the case of voice input, voice recognition software converts the voice into text.

[0159] Step 2:

[0160] The server analyzes the user's input using a natural language processing engine and extracts keywords.

[0161] Input: User's question data (in text form)

[0162] Output: Extracted keywords (e.g., "inventory status", "latest")

[0163] Specific operation: The server uses a natural language processing library such as spaCy to analyze the text and extract important keywords. For example, it performs morphological analysis to identify important words such as nouns and verbs.

[0164] Step 3:

[0165] The server searches the database based on the extracted keywords and obtains relevant documents and materials.

[0166] Input: Extracted keywords (e.g., "inventory status", "latest")

[0167] Output: Relevant documents and materials (e.g., documents containing the latest inventory information)

[0168] Specific operation: The server uses a database management system such as SQLite to search for documents and materials that match the keywords. For example, it executes an SQL query to obtain records that match the keywords.

[0169] Step 4:

[0170] The server sends the search results to the smartphone and displays them to the user.

[0171] Input: Relevant documents and materials (e.g., documents containing the latest inventory information)

[0172] Output: Search results displayed on the smartphone

[0173] Specific operation: The server converts the search results into a data format such as JSON and sends them to the smartphone. The smartphone application analyzes the received data and displays it in a user-friendly format.

[0174] Step 5:

[0175] The user checks the search results displayed on the smartphone and obtains the necessary information.

[0176] Input: Search results displayed on the smartphone

[0177] Output: Necessary information obtained by the user (e.g., the latest inventory information)

[0178] Specific operations: The user checks the search results displayed on the smartphone screen and obtains the necessary information. For example, check information such as inventory status and delivery schedule.

[0179] (Example 2)

[0180] Next, Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0181] Employees are required to quickly and efficiently search for necessary documents and materials and provide results sorted in descending order of relevance. However, in the conventional system, there is a problem that the search results are not properly sorted, and it takes time to find the necessary information. In addition, since the display format of the search results is inappropriate, there is a problem that it is difficult for the user to quickly find the necessary information.

[0182] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0183] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for sorting the search results in descending order of relevance, and means for generating the search results in a list format. As a result, it becomes possible for employees to quickly and efficiently find the necessary information.

[0184] The "natural language processing engine" is software for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[0185] The "means for searching" is a function for searching for documents and materials in the database based on the questions and keywords of employees.

[0186] The "means for providing" is a function for displaying the search results to employees.

[0187] The "sorting means" is a function for sorting search results in descending order of relevance.

[0188] The "generating means" is a function for formatting search results in a list format.

[0189] The "displaying means" is a function for displaying the generated search results on the terminals of employees.

[0190] The "machine learning model" is an algorithm for analyzing questions and keywords based on data.

[0191] The "categorizing means" is a function for classifying search results based on specific criteria and automatically storing them in folders or directories corresponding to each category.

[0192] Modes for Carrying Out the Invention

[0193] This invention is a system that enables employees to quickly and efficiently search for necessary documents and materials and provides search results sorted in descending order of relevance. The following describes specific embodiments of this system.

[0194] Configuration of the System

[0195] This system consists of three main elements: a server, terminals, and users. The server analyzes search queries using a natural language processing engine and a machine learning model, and searches for relevant documents and materials. The terminal is a device for users to input search queries and display search results. The user is an employee who uses the system to search for necessary information.

[0196] Hardware and Software to be Used

[0197] Server: A computer system equipped with a high-performance processor and a large amount of memory is used. Python (registered trademark)-based libraries (e.g., NLTK, spaCy) are installed on the server as a natural language processing engine. Also, ElasticSearch (registered trademark) is used for the search algorithm.

[0198] Terminal: A device such as a personal computer or smartphone used by the user. A web browser or a dedicated application is installed on the terminal.

[0199] Software: Machine learning models (e.g., BERT, GPT) are used for parsing search queries. This enables accurate parsing of the user's questions and keywords.

[0200] Operation of the System

[0201] 1. The user enters a search query

[0202] The user enters the necessary information in the search bar of the terminal. For example, enter "project report".

[0203] 2. The terminal sends the search query to the server

[0204] The terminal sends the search query entered by the user to the server. The query is sent using an HTTP request.

[0205] 3. The server receives the search query and queries Elasticsearch

[0206] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch.

[0207] 4. The server receives the search results from Elasticsearch

[0208] The server receives the search results returned from Elasticsearch. The search results contain information about relevant documents and materials.

[0209] 5. The server sorts the search results in descending order of relevance

[0210] The server sorts the received search results in descending order of relevance. The scoring function of Elasticsearch is used to evaluate the relevance.

[0211] 6. The server generates the search results in a list format

[0212] The server formats the sorted search results into a list in HTML format.

[0213] 7. The server sends the generated search results to the terminal

[0214] The server sends the generated search results in HTML format to the terminal. The results are sent using an HTTP response.

[0215] 8. The terminal displays the search results to the user

[0216] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format.

[0217] 9. The user selects the necessary documents and materials

[0218] The user selects the necessary documents and materials from the displayed search results. Click on the selected document to display the details.

[0219] Specific example

[0220] As a specific example, consider the case where a user searches for a "project report". The user enters "project report" from the terminal and clicks the search button. The server receives this query and searches for documents in the database using Elasticsearch. The search results are sorted in descending order of relevance and displayed to the user in a list format. The user can select the necessary document from this list.

[0221] Examples of prompt sentences

[0222] Examples of prompt sentences may include the following.

[0223] Prompt sentence: "Please search for the project report."

[0224] When this prompt sentence is input into the generative AI model, the AI model provides search results for quickly finding the project report required by the employee.

[0225] The flow of the specific process in Example 2 will be described with reference to FIG. 13.

[0226] Step 1:

[0227] The user inputs a search query.

[0228] The user inputs the necessary information into the search bar of the terminal. For example, the user inputs "project report". The input query is temporarily stored in the memory of the terminal.

[0229] Step 2:

[0230] The terminal sends the search query to the server.

[0231] The terminal sends the search query input by the user to the server as an HTTP request. This request contains the query input by the user. After sending the request, the terminal waits for a response from the server.

[0232] Step 3:

[0233] The server receives a search query and queries Elasticsearch.

[0234] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch. Specifically, the server passes the search query to the Elasticsearch API and instructs it to search for relevant documents and materials. The input is the search query, and the output is the search results from Elasticsearch.

[0235] Step 4:

[0236] The server receives search results from Elasticsearch.

[0237] The server receives the search results returned by Elasticsearch. The search results contain information on relevant documents and materials. The server saves this result in memory in preparation for the next processing. The input is the search results from Elasticsearch, and the output is the search results saved in the server's memory.

[0238] Step 5:

[0239] The server sorts the search results in descending order of relevance.

[0240] The server sorts the received search results in descending order of relevance using Elasticsearch's scoring function. Specifically, it calculates the relevance score for each document and sorts them in descending order. The input is the search results saved in the server's memory, and the output is the sorted search results.

[0241] Step 6:

[0242] The server generates the search results in a list format.

[0243] The server formats the sorted search results into a list in HTML format. Specifically, it encloses the title and summary of each document with HTML tags to make it in a user-friendly format. The input is the sorted search results, and the output is the search results in HTML format.

[0244] Step 7:

[0245] The server sends the search results generated to the terminal.

[0246] The server sends the generated search results in HTML format to the terminal as an HTTP response. The input is the search results in HTML format, and the output is the search results sent to the terminal.

[0247] Step 8:

[0248] The terminal displays the search results to the user.

[0249] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format. Specifically, the web browser parses the HTML and displays it on the screen. The input is the search results in HTML format received from the server, and the output is the search results displayed to the user.

[0250] Step 9:

[0251] The user selects the necessary documents and materials.

[0252] The user selects the necessary documents and materials from the displayed search results. By clicking on the selected document, the details are displayed. The input is the click operation of the user, and the output is the detailed display of the selected document.

[0253] (Application Example 2)

[0254] Next, Application Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0255] In a logistics center, employees are required to quickly search for necessary inventory information and delivery information and efficiently perform their operations. However, in the conventional system, there was a problem that it took time to search for information, and a large amount of irrelevant information was displayed, resulting in a decrease in work efficiency. Furthermore, since information search using smartphones was not fully utilized, there was also a problem that it was difficult to respond quickly on-site.

[0256] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 2 is realized by the following respective means.

[0257] In this invention, the server includes means for searching for relevant documents and materials from an employee's question or keyword using a natural language processing engine, means for providing the search results to the employee, means for organizing the documents and materials based on the search results, means for displaying inventory information and delivery information in descending order of relevance, and means for enabling the employee to quickly search for necessary information using a smartphone. As a result, the employee can quickly search for necessary information and improve work efficiency.

[0258] The "natural language processing engine" is software for analyzing an employee's question or keyword and searching for relevant documents and materials.

[0259] The "means for providing search results" is a function for displaying the searched documents and materials to the employee.

[0260] The "means for organizing documents and materials" is a function for categorizing documents and materials based on the search results and automatically storing them in folders or directories corresponding to each category.

[0261] The "means for displaying inventory information and delivery information in descending order of relevance" is a function for sorting and displaying inventory data and delivery data in descending order of relevance based on a search query.

[0262] "A means for employees to quickly search for necessary information using a smartphone" refers to a function that enables employees to quickly search for necessary information using a smartphone.

[0263] The system for implementing this invention is designed to enable employees in a logistics center to quickly search for necessary inventory information and delivery information. The following describes the specific embodiments of this system.

[0264] Configuration of the System

[0265] This system is composed of the following main components:

[0266] 1. Server: Equipped with a natural language processing engine and a machine learning model to analyze employees' questions and keywords.

[0267] 2. Smartphone: The terminal used by employees to input search queries and display search results.

[0268] 3. Database: Stores inventory information and delivery information.

[0269] Program Processing

[0270] The server operates as follows:

[0271] 1. Natural language processing engine: Analyzes questions and keywords input by employees from the smartphone and searches for relevant documents and materials.

[0272] 2. Machine learning model: Generates search results in descending order of relevance based on the analysis of questions and keywords.

[0273] 3. Provision of search results: Sends the search results to the smartphone and provides them to employees.

[0274] 4. Data Sorting: Categorize documents and materials based on search results and automatically store them in the corresponding folders or directories for each category.

[0275] 5. Display of Inventory Information and Delivery Information: Sort and display inventory data and delivery data in descending order of relevance.

[0276] Hardware and Software to be Used

[0277] Hardware: Smartphones, Servers

[0278] Software: Python, Pandas, Scikit-learn, Natural Language Processing Engine (e.g., SpaCy)

[0279] Specific Examples

[0280] When an employee searches for "products with insufficient inventory", the system operates as follows:

[0281] 1. The employee enters "products with insufficient inventory" into the smartphone.

[0282] 2. The natural language processing engine on the server analyzes this query and searches for relevant inventory information.

[0283] 3. The machine learning model generates search results in descending order of relevance.

[0284] 4. The search results are displayed on the smartphone, and the employee can view the list of products with insufficient inventory.

[0285] Examples of Prompt Sentences

[0286] Search Query: "products with insufficient inventory"

[0287] In this way, employees in the logistics center can quickly search for the necessary information and improve work efficiency.

[0288] The flow of the specific process in Application Example 2 will be described with reference to FIG. 14.

[0289] Step 1:

[0290] The user enters a search query into the smartphone.

[0291] Input: The user enters "out-of-stock products" into the smartphone.

[0292] Output: The search query is sent to the server.

[0293] Specific operation: The user enters "out-of-stock products" into the search bar of the smartphone and presses the search button.

[0294] Step 2:

[0295] The server analyzes the search query using a natural language processing engine.

[0296] Input: The search query "out-of-stock products" sent from the smartphone.

[0297] Output: The analyzed query data.

[0298] Specific operation: The natural language processing engine (e.g., SpaCy) of the server tokenizes the search query and extracts important keywords.

[0299] Step 3:

[0300] The server searches for relevant documents and materials using a machine learning model.

[0301] Input: The analyzed query data.

[0302] Output: A list of relevant documents and materials.

[0303] Specific operation: The server's machine learning model (e.g., Scikit-learn) uses the analyzed query data to search for inventory information and delivery information in the database and sorts them in descending order of relevance.

[0304] Step 4:

[0305] The server sends the search results to the smartphone.

[0306] Input: A list of relevant documents and materials.

[0307] Output: Search results displayed on the smartphone.

[0308] Specific operation: The server sends a list of relevant documents and materials to the smartphone and displays them to the user.

[0309] Step 5:

[0310] The user checks the search results on the smartphone.

[0311] Input: Search results displayed on the smartphone.

[0312] Output: The user checks the required information.

[0313] Specific operation: The user scrolls through the search results displayed on the smartphone screen and checks the required inventory information and delivery information.

[0314] Step 6:

[0315] The server organizes the documents and materials based on the search results.

[0316] Input: A list of relevant documents and materials.

[0317] Output: Categorized documents and materials.

[0318] Specific operation: The server categorizes documents and materials based on the search results and automatically stores them in the folders or directories corresponding to each category.

[0319] Step 7:

[0320] The server displays the inventory information and delivery information in descending order of relevance.

[0321] Input: Relevant inventory information and delivery information.

[0322] Output: Inventory information and delivery information sorted in descending order of relevance.

[0323] Specific operation: The server sorts the inventory data and delivery data in descending order of relevance and displays it to the user.

[0324] (Example 3)

[0325] Next, Example 3 of Form Example 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0326] There is a problem that it is difficult for employees to efficiently search for and organize the necessary information. In particular, when there are a large number of documents and materials, manual search and organization require time and effort, resulting in a decrease in work efficiency. Also, when the search results are not properly categorized, it is difficult to quickly find the necessary information.

[0327] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for searching for relevant information from an employee's question or keyword using a natural language processing engine, means for providing the search results to the employee, means for analyzing and categorizing the information based on the search results, and means for automatically storing the categorized information in the folders or directories corresponding to each category. Thereby, employees can efficiently search for and organize the necessary information.

[0328] The "Natural Language Processing Engine" is software for analyzing natural language and understanding its meaning.

[0329] An "employee" is an individual who belongs to a company or organization and performs work.

[0330] "Questions and Keywords" are input data used by employees when searching for information.

[0331] "Related Information" refers to documents and materials retrieved based on an employee's questions and keywords.

[0332] "Means of Searching" are methods and technologies for obtaining related information based on an employee's questions and keywords.

[0333] "Means of Providing" are methods and technologies for displaying search results to employees.

[0334] "Means of Analyzing and Categorizing" are methods and technologies for analyzing search results and classifying them into specific categories.

[0335] "Folders and Directories" are virtual storage locations for organizing information within a computer.

[0336] "Means of Automatically Storing" are methods and technologies for automatically moving analyzed information to corresponding folders and directories.

[0337] "Machine Learning Model" is an algorithm for learning based on data and performing prediction and classification.

[0338] This invention is a system for employees to efficiently search for and organize necessary information. The specific embodiments of this system will be described below.

[0339] Configuration of the System

[0340] This system consists of three main elements: a server, a terminal, and a user.

[0341] 1. Server:

[0342] The server uses a natural language processing engine to search for relevant information from employees' questions and keywords. Specifically, it uses search engines such as Apache (registered trademark) Solr or Elasticsearch. Additionally, it uses natural language processing tools such as Google (registered trademark) Cloud Natural Language API or IBM Watson (registered trademark) Natural Language Understanding to analyze the search results and categorize them. Furthermore, it uses Python's os module and shutil module to automatically store the categorized information in the corresponding folders or directories.

[0343] 2. Terminal:

[0344] The terminal is a device for users to input search queries and view search results. The terminal is composed of computer devices such as personal computers, tablets, and smartphones.

[0345] 3. User:

[0346] The user is an employee who uses the system to search for and organize information. The user inputs search queries using the terminal and views the organized information.

[0347] Operation of the system

[0348] 1. Input of search query by user:

[0349] The user inputs questions or keywords into the search bar of the terminal. For example, input "2023 annual business report".

[0350] 2. Obtaining of search results by server:

[0351] The server receives the search query entered by the user and retrieves relevant information from the database using Apache Solr or Elasticsearch.

[0352] 3. Analysis and Categorization of Search Results by the Server:

[0353] The server analyzes the retrieved search results using the Google Cloud Natural Language API or IBM Watson Natural Language Understanding and classifies them into appropriate categories. For example, it classifies them into categories such as "Business Report", "Financial Report", "Market Analysis", etc.

[0354] 4. Storage of Information in Folders for Each Category by the Server:

[0355] The server automatically stores the categorized information in the folders or directories corresponding to each category using Python's os module or shutil module. For example, the information classified into the "Business Report" category is stored in the "Business Report" folder.

[0356] 5. Confirmation of Organized Information by the User:

[0357] The user checks the organized folders using the terminal. For example, the user can open the "Business Report" folder and view the necessary information.

[0358] Examples of Specific Examples and Prompt Sentences

[0359] Specific Example:

[0360] The specific actions when the user searches for the "Business Report for 2023" are as follows.

[0361] 1. User: Enter "2023 Business Report" in the search bar of the terminal and click the search button.

[0362] 2. Server: Receive the search query and use Apache Solr to retrieve relevant information from the database.

[0363] 3. Server: Analyze the retrieved information using the Google Cloud Natural Language API and classify it into the "Business Report" category.

[0364] 4. Server: Move the classified information to the "Business Report" folder using the Python os module.

[0365] 5. User: Open the "Business Report" folder on the terminal and check the necessary information.

[0366] Example of a prompt sentence:

[0367] "Search for the 2023 Business Report and automatically store the relevant information in the Business Report folder."

[0368] With this system, users can efficiently search for and organize the necessary information. The flow of the specific process in Example 3 will be described with reference to Figure 15.

[0369] Step 1: Input of the search query by the user

[0370] The user enters a question or keyword in the search bar of the terminal. For example, enter "2023 Business Report". The input data is text related to the information the user wants to search for. The output is sent to the server as a search query.

[0371] Step 2: Acquisition of the search results by the server

[0372] The server receives the search query entered by the user and uses a search engine (e.g., Apache Solr or Elasticsearch) to retrieve relevant information from the database. The input data is the user's search query. The server sends the search query to the search engine and retrieves relevant documents and materials. The output is a list of documents and materials as search results.

[0373] Step 3: Analysis and Categorization of Search Results by the Server

[0374] The server analyzes the retrieved search results using a natural language processing engine (e.g., Google Cloud Natural Language API or IBM Watson Natural Language Understanding). The input data is the documents and materials obtained as search results. The server analyzes these documents and classifies them into appropriate categories. For example, it classifies them into categories such as "Business Report", "Financial Report", "Market Analysis", etc. The output is a list of categorized documents and materials.

[0375] Step 4: Storage of Categorized Documents into Folders by the Server

[0376] The server automatically stores the categorized documents and materials into folders or directories corresponding to each category. The input data is the categorized documents and materials. The server uses the os module or shutil module in Python to obtain the document paths and move them to the corresponding folders. For example, the documents classified into the "Business Report" category are stored in the "Business Report" folder. The output is the documents and materials stored in the folders.

[0377] Step 5: Confirmation of Organized Information by the User

[0378] The user checks the folders organized using the terminal. The input data is the documents and materials stored in the folders. The user can open, for example, the "Sales Report" folder and view the necessary documents. The output is the documents and materials viewed by the user.

[0379] With this system, the user can efficiently search for and organize the necessary information.

[0380] (Application Example 3)

[0381] Next, Application Example 3 of Morphological Example 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0382] In conventional document management systems, the search and organization of documents and materials are often performed manually, which has the problem of low efficiency. Also, at sites such as logistics centers, there are many paper-based documents, and means for digitizing and efficiently managing them have been demanded. Furthermore, the categorization of documents and the upload to cloud storage are often also performed manually, which has the issue of taking time and effort.

[0383] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following respective means.

[0384] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for extracting text from images using optical character recognition technology, means for categorizing documents based on the extracted text and automatically storing them in corresponding folders, and means for uploading documents to cloud storage. Thereby, efficient search, organization, digitization, and automatic upload to cloud storage of documents and materials become possible.

[0385] A "natural language processing engine" is software for analyzing natural language and retrieving relevant information from questions and keywords.

[0386] "Employees" refer to people who belong to a company or organization and perform their duties.

[0387] "Means of retrieval" refers to methods and techniques for finding specific information.

[0388] "Means of provision" refers to methods and techniques for displaying search results to users or making them accessible.

[0389] "Means of sorting" refers to methods and techniques for classifying and managing documents and materials based on specific criteria.

[0390] "Optical character recognition technology" is a technology for extracting character information from images.

[0391] "Image" refers to a digital file containing visual information.

[0392] "Text" refers to digital data containing character information.

[0393] "Categorization" refers to classifying information based on specific criteria.

[0394] "Folder" refers to a virtual container for organizing digital data.

[0395] "Cloud storage" refers to a remote server for storing data over the Internet.

[0396] "Upload" refers to transferring data from a local device to a remote server.

[0397] The system for implementing this invention is configured as follows. First, the server uses a natural language processing engine to search for relevant documents and materials from employees' questions and keywords. The natural language processing engine analyzes the questions and keywords based on a machine learning model. The analyzed results are provided to the employees.

[0398] Next, the server organizes the documents and materials based on the search results. As a means of organization, optical character recognition technology (OCR) is used to extract text from images. The extracted text is categorized based on specific keywords and automatically stored in the corresponding folders. Furthermore, the documents are uploaded to cloud storage.

[0399] To implement this system, the following hardware and software are used. As hardware, a smartphone with a camera is required. As software, Python, an OCR library, PIL (Python Imaging Library), and Google Cloud Storage are used.

[0400] As a specific example, consider a document organization application used in a logistics center. When an employee takes a photo of a shipping instruction document with the camera of a smartphone, the image is converted into text using OCR technology. The converted text is classified into the category of "shipping instruction document" and automatically stored in the corresponding folder. After that, the document is uploaded to cloud storage.

[0401] Examples of prompt texts can be as follows.

[0402] "Please develop a document organization application used in a logistics center. It is an application that analyzes document images taken with the camera of a smartphone using OCR technology, automatically classifies them into categories such as shipping instruction documents, receiving documents, and inventory lists, and has the function of uploading them to cloud storage."

[0403] In this way, efficient search, organization, digitization of documents and materials, and automatic upload to cloud storage become possible.

[0404] The flow of the specific process in Application Example 3 will be described with reference to FIG. 16.

[0405] Step 1:

[0406] The user takes a picture of a document with the camera of a smartphone.

[0407] Input: Paper-based document

[0408] Output: Digital image file

[0409] Specific operation: The user launches the camera app of the smartphone and takes a picture of the document. The captured image is saved in the smartphone.

[0410] Step 2:

[0411] The terminal extracts text from the image using optical character recognition technology (OCR).

[0412] Input: Digital image file

[0413] Output: Extracted text data

[0414] Specific operation: The terminal inputs the saved image file into OCR software and extracts the character information in the image as text data.

[0415] Step 3:

[0416] The terminal analyzes the extracted text and categorizes it.

[0417] Input: Extracted text data

[0418] Output: Category information

[0419] Specific operation: The terminal analyzes the extracted text data and determines the category based on specific keywords (e.g., "shipping instruction", "receipt", etc.).

[0420] Step 4:

[0421] The terminal automatically stores the document in the corresponding folder based on the category.

[0422] Input: Category information, digital image file

[0423] Output: Document stored in the folder

[0424] Specific operation: The terminal moves or copies the digital image file to the corresponding folder based on the determined category.

[0425] Step 5:

[0426] The terminal uploads the document to the cloud storage.

[0427] Input: Document stored in the folder

[0428] Output: Document saved in the cloud storage

[0429] Specific operation: The terminal uploads the document in the folder to the cloud storage service (Google Cloud Storage). When the upload is completed, the document is saved on the cloud.

[0430] Step 6:

[0431] The server uses the natural language processing engine to search for relevant documents and materials from the employees' questions and keywords.

[0432] Input: Employees' questions and keywords

[0433] Output: Search results of relevant documents and materials

[0434] Specific operation: The server inputs the questions or keywords entered by the employee into the natural language processing engine and searches for relevant documents and materials. The search results are provided to the employee.

[0435] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[0436] "Form Example 1"

[0437] In one embodiment of the present invention, a natural language processing engine and an emotion engine are combined. When an employee inputs a question or keyword to the system, the natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. At the same time, the emotion engine analyzes the emotional context of the employee's question or keyword. For example, when an employee inputs the phrase "I can't find the latest sales report", the emotion engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[0438] "Form Example 2"

[0439] In another embodiment of the present invention, the emotion engine sorts documents and materials based on the employee's emotional context. Specifically, the emotion engine analyzes the employee's emotional context and automatically stores the documents and materials in the folders or directories corresponding to each emotional context. For example, when an employee inputs the phrase "I can't find the sales report", the emotion engine recognizes the employee's frustration from this phrase, and as a result, the system automatically stores the relevant documents and materials in the "frustration" folder.

[0440] "Form Example 3"

[0441] In one embodiment of the present invention, a natural language processing engine and an emotion engine are combined. When an employee inputs a question or keyword to the system, the natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. At the same time, the emotion engine analyzes the emotional context of the employee's question or keyword. For example, when an employee inputs the phrase "I can't find the latest sales report", the emotion engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[0442] The processing flow of each exemplary embodiment will be described below.

[0443] "Exemplary Embodiment 1"

[0444] Step 1: The employee inputs a question or keyword to the system.

[0445] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database.

[0446] Step 3: The emotion engine analyzes the emotional context of the employee's question or keyword.

[0447] Step 4: The natural language processing engine uses the analysis result of the emotion engine to provide more relevant documents and materials.

[0448] "Exemplary Embodiment 2"

[0449] Step 1: The employee inputs a question or keyword to the system.

[0450] Step 2: The emotion engine analyzes the emotional context of the employee's question or keyword.

[0451] Step 3: Based on the analysis results of the emotion engine, automatically store documents and materials in folders and directories corresponding to each emotional context.

[0452] (Example 1)

[0453] Next, Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0454] In a conventional system, when searching for relevant documents and materials for questions and keywords input by employees, the emotional context cannot be considered, and it is difficult to fully meet the needs of employees. Also, in organizing and providing search results, appropriate information reflecting the emotions of employees is not provided, so efficient business operations are hindered.

[0455] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0456] In this invention, the server includes means for searching for relevant documents and materials from questions and keywords of employees using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for analyzing the emotional context of questions and keywords of employees using an emotion engine, and means for providing highly relevant documents and materials based on the emotional context. Thereby, appropriate information provision considering the emotions of employees becomes possible, and it becomes possible to improve work efficiency and employee satisfaction.

[0457] The "natural language processing engine" is software for analyzing questions and keywords input by employees, understanding their meanings, and searching for relevant documents and materials.

[0458] A "machine learning model" is an algorithm that learns based on data to find patterns and rules, and is used to improve the analysis ability of a natural language processing engine.

[0459] An "emotion engine" is software that analyzes the emotional context contained in employees' questions and keywords and recognizes that emotion.

[0460] "Documents and materials" is a general term for texts, reports, presentations, datasheets, etc. that contain information necessary for employees to perform their work.

[0461] A "database" is an information system for efficiently storing, searching, and managing documents and materials.

[0462] "Search results" is a list of documents and materials obtained from a database based on questions and keywords analyzed by a natural language processing engine and an emotion engine.

[0463] "Categorization" is a process of classifying documents and materials according to specific themes or attributes based on search results.

[0464] "Folders and directories" are virtual storage locations on a computer for organizing and storing documents and materials.

[0465] An "employee" is a user who inputs questions and keywords to perform work using the system.

[0466] Mode for Carrying Out the Invention

[0467] This invention is a system that analyzes questions and keywords input by employees and provides relevant documents and materials. The system combines a natural language processing engine and an emotion engine to provide information considering the emotional context of employees.

[0468] Hardware and Software to be Used

[0469] Server

[0470] The server performs the main processing of the system using the following software:

[0471] Natural Language Processing Engine (e.g., BERT, GPT-3)

[0472] Sentiment Engine (e.g., Sentiment Analysis API)

[0473] Database Management System (e.g., MySQL, PostgreSQL)

[0474] Terminal

[0475] The terminal provides an interface for employees to enter questions and keywords. The terminal uses the following software:

[0476] Web Browser or Dedicated Application

[0477] HTTP Client Library

[0478] User

[0479] The user (employee) enters questions and keywords into the system and checks the search results.

[0480] Data Processing and Data Calculation

[0481] Natural Language Processing Engine

[0482] The server passes the questions and keywords entered by the employee to the natural language processing engine for analysis. The natural language processing engine uses a machine learning model (e.g., BERT, GPT-3) to understand the meaning of the input text and extracts keywords for searching relevant documents and materials.

[0483] Emotion Engine

[0484] The server passes the input text to the emotion engine and analyzes the emotional context. The emotion engine recognizes the emotions of employees (e.g., frustration, joy) from the input text and provides relevant documents and materials based on that information.

[0485] Database Search

[0486] The server searches the database for relevant documents and materials based on the analysis results of the natural language processing engine and the emotion engine. The search results are prioritized considering the emotional context of the employees.

[0487] Provision and Arrangement of Results

[0488] The server sends the search results to the terminal, and the terminal displays them to the user. Furthermore, the documents and materials are categorized based on the search results and automatically stored in the corresponding folders or directories for each category.

[0489] Specific Example

[0490] When an employee enters "I can't find the latest sales report", the terminal sends this input to the server. The server uses the natural language processing engine to analyze the keyword "latest sales report" and the emotion engine to recognize the employee's frustration from the phrase "can't find". The server searches the database for relevant documents based on this information and sends the search results to the terminal. The terminal displays the search results to the user.

[0491] Example of Prompt Sentence

[0492] "Please explain how the system responds when an employee enters 'I can't find the latest sales report' into the system."

[0493] By inputting this prompt text into the generation AI model, a response that explains the operation of the system in detail can be obtained.

[0494] The flow of the specific process in Example 1 will be described with reference to FIG. 17.

[0495] Step 1:

[0496] The user inputs a question or keyword.

[0497] The user inputs a question or keyword such as "The latest business report cannot be found" into the system interface. The input text is displayed in the input field of the terminal. As a specific operation, the user uses the keyboard to input text and clicks the send button.

[0498] Step 2:

[0499] The terminal sends the input to the server.

[0500] The terminal sends the text input by the user to the server. The input text is sent to the server as an HTTP request. As a specific operation, the application on the terminal generates an HTTP request and sends it to the server.

[0501] Step 3:

[0502] The server analyzes the input with a natural language processing engine.

[0503] The server passes the received text to the natural language processing engine for analysis. By analyzing the input text, relevant keywords and phrases are extracted. As a specific operation, the server calls the API of the natural language processing engine to obtain the analysis result. The input is the user's text, and the output is the analyzed keywords and phrases.

[0504] Step 4:

[0505] The server analyzes the emotional context using the emotion engine.

[0506] The server passes the input text to the emotion engine to analyze the emotional context. The emotion engine recognizes the emotions (e.g., frustration) of the employees from the input text. Specifically, the server calls the API of the emotion engine to obtain the emotion analysis result. The input is the user's text, and the output is the analyzed emotion information.

[0507] Step 5:

[0508] The server searches the database for relevant documents and materials.

[0509] The server searches the database for relevant documents and materials based on the analysis results of the natural language processing engine and the emotion engine. The search results are prioritized considering the emotional context of the employees. Specifically, the server generates an SQL query and executes the search against the database. The input is the analysis result, and the output is a list of the retrieved documents and materials.

[0510] Step 6:

[0511] The server sends the search results to the terminal.

[0512] The server sends the search results to the terminal. The search results are sent to the terminal as an HTTP response. Specifically, the server generates an HTTP response and sends it to the terminal. The input is the search result, and the output is the data sent to the terminal.

[0513] Step 7:

[0514] The terminal displays the search results to the user.

[0515] The terminal displays the search results received from the server to the user, who can then check related documents and materials. Specifically, the terminal application displays the search results on the screen. The input is data from the server, and the output is the search results displayed to the user.

[0516] (Application example 1)

[0517] Next, a description will be given of Application Example 1 of Embodiment Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0518] It is important for employees to be able to quickly and accurately search for and obtain the documents and materials they need within the factory. However, with conventional systems, it often took employees a long time to find the information they needed, reducing efficiency. Furthermore, providing information without taking into account the emotional state of employees can increase stress and frustration. To solve these problems, support that takes into account the emotional context is needed, rather than simply analyzing employees' questions and keywords and providing relevant documents and materials.

[0519] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0520] In this invention, the server includes means for searching for relevant documents and materials based on the employee's questions and keywords using a natural language processing engine, means for providing the search results to the employee, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of the employee's questions and keywords using a sentiment analysis engine, and means for providing additional support to the employee based on the emotional context, thereby enabling the employee to not only quickly and accurately obtain the information they need, but also receive emotional support.

[0521] The "Natural Language Processing Engine" is software for analyzing employees' questions and keywords and searching for relevant documents and materials.

[0522] The "Machine Learning Model" is an algorithm that learns based on data and analyzes employees' questions and keywords.

[0523] The "Sentiment Analysis Engine" is software for analyzing the emotional context of employees' questions and keywords.

[0524] "Documents and materials" include information such as manuals, work instructions, and reports used within the factory.

[0525] The "Search Results" is a list of relevant documents and materials retrieved by the Natural Language Processing Engine.

[0526] "Categorization" is a process of classifying documents and materials into specific categories based on the search results.

[0527] "Folders and directories" are digital storage locations for organizing and storing documents and materials.

[0528] "Additional support" refers to encouraging words and supplementary information provided based on the emotional context analyzed by the Sentiment Analysis Engine.

[0529] To implement this invention, the following hardware and software are used. As hardware, a robot body, built-in microphone and speaker, and display used within the factory are required. As software, a natural language processing engine (e.g., spaCy, BERT), a sentiment analysis engine (e.g., IBM Watson Tone Analyzer), a database management system (e.g., MySQL, PostgreSQL), and robot control software (e.g., ROS - Robot Operating System) are used.

[0530] When an employee inputs a question or keyword to the robot by voice, the built-in microphone captures the voice and converts it into text using speech recognition software (e.g., Google Speech-to-Text API). Next, the natural language processing engine analyzes the converted text and extracts the keyword and the intention of the question. For example, when "Show me the latest work instructions" is input, the keyword "latest work instructions" is extracted.

[0531] After that, the sentiment analysis engine analyzes the emotional context of the text. For example, when "I can't find the work instructions" is input, frustration is detected. Next, the server searches the database for relevant documents and materials based on the extracted keyword. For example, it retrieves documents related to "latest work instructions" from the database.

[0532] The search results are displayed on the display and notified to the employee by voice. Based on the result of the sentiment analysis, encouraging words and additional support are provided if necessary. For example, a response such as "I can't find the work instructions and I'm in trouble. I'll search immediately, so please wait a moment." is possible.

[0533] As a specific example, the following examples of prompt sentences are shown.

[0534] Employee: "Show me the latest work instructions"

[0535] Robot: "Searching for the latest work instructions. Please wait a moment."

[0536] (After database search)

[0537] Robot: "Here is the latest work instructions. It is displayed on the screen."

[0538] Also, as specific examples of sentiment analysis, the following exchanges can be considered.

[0539] Employee: "I can't find the work instruction manual. What should I do?"

[0540] Robot: "You can't find the work instruction manual. You must be having a hard time. I'll search for it right away. Please wait a moment."

[0541] (After database search)

[0542] Robot: "This is the latest work instruction manual. I've displayed it on the screen. Is there anything else I can help you with?"

[0543] In this way, not only can employees quickly and accurately obtain the necessary information, but they can also receive emotional support.

[0544] The flow of the specific process in Application Example 1 will be described with reference to FIG. 18.

[0545] Step 1:

[0546] The user inputs questions or keywords to the robot by voice.

[0547] Input: User's voice input

[0548] Output: Voice data

[0549] Specific operation: The user says to the robot, "Show me the latest work instruction manual."

[0550] Step 2:

[0551] The built-in microphone of the robot captures the voice and converts it into text using speech recognition software (e.g., Google Speech-to-Text API).

[0552] Input: Voice data

[0553] Output: Text data

[0554] Specific operation: The robot records the user's voice and converts the voice data into text.

[0555] Step 3:

[0556] The server uses a natural language processing engine (e.g., spaCy, BERT) to analyze the converted text and extract keywords and the intent of the question.

[0557] Input: Text data

[0558] Output: Extracted keywords and intent

[0559] Specific operation: The server extracts the keyword "latest work instruction" from the text "Show me the latest work instruction".

[0560] Step 4:

[0561] The server uses a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the emotional context of the text.

[0562] Input: Text data

[0563] Output: Sentiment analysis result

[0564] Specific operation: The server detects frustration from the text "The work instruction cannot be found".

[0565] Step 5:

[0566] The server searches the database for relevant documents and materials based on the extracted keywords.

[0567] Input: Extracted keywords

[0568] Output: Search results (relevant documents and materials)

[0569] Specific operation: The server searches the database using the keyword "latest work instruction" and retrieves the relevant documents.

[0570] Step 6:

[0571] The server displays the search results on the display and notifies the user by voice.

[0572] Input: Search results

[0573] Output: Display on the display and voice notification

[0574] Specific operation: The server notifies the user by voice with "This is the latest work instruction. It has been displayed on the screen." and displays the document on the display.

[0575] Step 7:

[0576] Based on the result of sentiment analysis, the server provides encouraging words or additional support as needed.

[0577] Input: Sentiment analysis result

[0578] Output: Additional support (such as encouraging words)

[0579] Specific operation: The server notifies the user by voice with "The work instruction cannot be found. You must be troubled. I will search immediately, so please wait a moment."

[0580] (Example 2)

[0581] Next, Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0582] There is a problem that it is difficult for employees to quickly and efficiently find the necessary documents and materials. In addition, since information is provided without considering the emotional context of employees, there is a possibility that employees' stress and frustration will increase. Furthermore, since there is a lack of means to organize documents and materials based on the emotions of employees, information management may become complicated.

[0583] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in the second embodiment is realized by the following means.

[0584] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion analysis engine, and means for automatically storing documents and materials in specific folders or directories based on the emotional context. As a result, it becomes possible for employees to quickly and efficiently find the necessary information, and by organizing information based on the emotional context, the stress and frustration of employees can be reduced.

[0585] The "natural language processing engine" is software for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[0586] The "machine learning model" is an algorithm that learns patterns based on data and performs prediction and classification.

[0587] The "emotion analysis engine" is software for analyzing the emotional context from the text input by employees.

[0588] "Folders and directories" are digital storage locations for organizing and storing documents and materials.

[0589] "Search Results" refer to a list of relevant documents and materials returned by a search engine based on employees' questions or keywords.

[0590] "Employee" refers to a user who searches and organizes documents and materials using the system.

[0591] "Documents and materials" refer to digital files containing information necessary for employees in their work.

[0592] "Emotional context" refers to the emotions and psychological states that can be read from the text input by employees.

[0593] This invention is a system for employees to quickly and efficiently find the necessary documents and materials. The system consists of a server including a natural language processing engine, an emotion analysis engine, a search engine, and a database, and a terminal operated by the user.

[0594] The server receives the search query input by the employee from the terminal and analyzes it using the natural language processing engine. The natural language processing engine incorporates a machine learning model to understand the employee's questions and keywords and search for relevant documents and materials. The search engine uses general search software such as Apache Solr or Elasticsearch.

[0595] The search results are displayed in a list format in descending order of relevance. The server provides the search results to the user's terminal, and the user can select the necessary documents and materials.

[0596] Furthermore, the server uses a sentiment analysis engine to analyze the emotional context of employees. The sentiment analysis engine uses sentiment analysis APIs such as IBM Watson and Microsoft® Azure®. It recognizes the sentiment from the phrases or search queries input by employees and organizes documents and materials based on it. For example, when an employee inputs a phrase like "Can't find the sales report", the sentiment analysis engine recognizes frustration and automatically stores the relevant documents and materials in the "Frustration" folder.

[0597] As a specific example, consider the case where a user inputs "Looking for marketing materials for the new product". The terminal sends this query to the server. The server uses Elasticsearch to search for relevant marketing materials and generates results in descending order of relevance. At the same time, the server uses the sentiment analysis API of IBM Watson to analyze the sentiment of "expectation". The server displays the search results in a list format on the user's terminal and automatically stores the relevant documents in the "Expectation" folder. The user can select and view the required materials from the displayed list.

[0598] Examples of prompt sentences are as follows:

[0599] "Looking for marketing materials for the new product"

[0600] "Can't find the sales report"

[0601] "Looking for materials summarizing customer feedback"

[0602] In this way, the server can quickly and efficiently provide the information required by the user.

[0603] The flow of the specific process in Example 2 will be described with reference to FIG. 19.

[0604] Step 1:

[0605] The user enters a search query.

[0606] The user enters the necessary information in the search bar of the terminal. For example, enter "Looking for marketing materials for new products". The entered search query becomes the input data for the next step.

[0607] Step 2:

[0608] The terminal sends the search query to the server.

[0609] The terminal sends the search query entered by the user to the server. An HTTP request is used for this communication. The sent search query becomes the input data for the server.

[0610] Step 3:

[0611] The server generates search results using a search engine.

[0612] The server passes the received search query to a search engine such as Apache Solr or Elasticsearch to search for relevant documents and materials. The search engine returns highly relevant results from the indexed database. The search results become the input data for the next step.

[0613] Step 4:

[0614] The server analyzes the user's sentiment using a sentiment analysis engine.

[0615] The server uses sentiment analysis APIs such as IBM Watson or Microsoft Azure to analyze the sentiment from the user's search query. For example, recognize sentiments such as "expectation" or "excitement" from the query "Looking for marketing materials for new products". The analyzed sentiment becomes the input data for the next step.

[0616] Step 5:

[0617] The server displays the search results in a list format in descending order of relevance.

[0618] The server sorts the results returned by the search engine in descending order of relevance and displays them to the user's terminal in a list format. The user can select the necessary documents and materials from this list. The displayed search results become the input data for the next step.

[0619] Step 6:

[0620] The server organizes documents and materials based on sentiment.

[0621] The server automatically stores the relevant documents and materials in specific folders or directories based on the sentiment analyzed by the sentiment analysis engine. For example, if the sentiment of "expectation" is recognized, the relevant documents are stored in the "expectation" folder. The organized documents and materials become the input data for the next step.

[0622] Step 7:

[0623] The user selects the necessary documents and materials.

[0624] The user selects the necessary documents and materials from the displayed search result list. The selected documents and materials are either downloaded to the user's terminal or displayed in a viewable state. The selected documents and materials become the final output data.

[0625] (Application Example 2)

[0626] Next, Application Example 2 of Morphological Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0627] If employees cannot quickly find the necessary documents and materials, there are problems such as a decline in work efficiency and an increase in stress. In addition, since appropriate support according to the emotional state of employees is not provided, the quality of work may decline. Especially at the work site such as a factory, it is required that work instructions and manuals be searched quickly and accurately, but this is not sufficiently achieved by the current system.

[0628] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following respective means.

[0629] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to the employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion engine and organizing the documents and materials based on it, and means for sorting relevant documents and materials according to the emotions of employees and providing detailed explanations and videos if necessary. Thereby, employees can quickly find the necessary information, and appropriate support according to the emotional state is provided, so that it is possible to improve work efficiency and reduce stress.

[0630] The "natural language processing engine" is a technology for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[0631] The "search results" are a list of documents and materials searched by the natural language processing engine.

[0632] The "emotion engine" is a technology for analyzing the emotional context of employees and organizing documents and materials based on it.

[0633] The "emotional context" is the emotional state inferred from the questions and keywords input by employees.

[0634] "Sorting" means arranging documents and materials based on specific criteria.

[0635] "Detailed explanation" means providing additional information and specific procedures for documents and materials.

[0636] "Video" means video content for providing information visually.

[0637] "Folder" means a virtual storage location for organizing documents and materials.

[0638] "Directory" means a structure for organizing files and folders in a computer system.

[0639] The system for implementing this invention has the following configuration. The server includes a natural language processing engine, an emotion engine, a document search engine, and a database. The terminal includes a voice input device, a display, and an interface. The user accesses the system through the terminal and inputs questions or keywords.

[0640] The server first uses the natural language processing engine to analyze the questions and keywords input by the user. Based on this analysis, the document search engine searches the database for relevant documents and materials. The search results are sorted in descending order of relevance and provided to the user.

[0641] Next, the emotion engine analyzes the emotional context from the user's input. For example, when the user inputs "I don't know how to install this part", the emotion engine recognizes the user's frustration. Based on this emotional context, the server further sorts the relevant documents and materials and provides detailed explanations and videos if necessary.

[0642] As a specific example, consider the case where a factory worker inputs "I don't know how to install this part". In this case, the server searches for relevant manuals, and if the emotion engine recognizes frustration, it provides detailed procedures or videos. As a result, the worker can quickly obtain the necessary information and work efficiency is improved.

[0643] Examples of prompt sentences to be input into the generative AI model are as follows:

[0644] Please create a program that analyzes the emotions of employees who input the phrase "I don't know how to install this part" and searches for and displays relevant documents and manuals. If the employee is feeling frustrated, also provide a detailed explanation or video.

[0645] The flow of the specific process in Application Example 2 will be described with reference to FIG. 20.

[0646] Step 1:

[0647] The user inputs questions or keywords through the terminal.

[0648] Input: The user inputs questions or keywords in voice or text.

[0649] Output: The terminal sends the input questions or keywords to the server.

[0650] Step 2:

[0651] The server analyzes the user's questions or keywords using a natural language processing engine.

[0652] Input: Questions or keywords sent from the terminal.

[0653] Data processing: The natural language processing engine analyzes the meaning of the questions or keywords using a machine learning model.

[0654] Output: Keywords and phrases as analysis results.

[0655] Step 3:

[0656] The server uses a document search engine to search the database for relevant documents and materials based on the analysis results.

[0657] Input: Analysis results of the natural language processing engine.

[0658] Data operation: The document search engine searches for documents and materials in the database and lists them in descending order of relevance.

[0659] Output: A list of relevant documents and materials.

[0660] Step 4:

[0661] The server sends the search results to the terminal, and the terminal provides them to the user.

[0662] Input: A list of relevant documents and materials.

[0663] Output: The terminal displays the search results to the user.

[0664] Step 5:

[0665] The server uses an emotion engine to analyze the emotional context from the user's input.

[0666] Input: The user's question or keywords.

[0667] Data processing: The emotion engine analyzes the emotion from the user's input and identifies the emotional context.

[0668] Output: Emotional context (e.g., frustration).

[0669] Step 6:

[0670] The server resorts relevant documents and materials based on the emotional context, and adds detailed explanations and videos if necessary.

[0671] Input: A list of documents and materials related to the emotional context.

[0672] Data operation: Resort documents and materials based on the emotional context, and add detailed explanations and videos.

[0673] Output: A list of resorted documents and materials and additional explanations and videos.

[0674] Step 7:

[0675] The server sends the list of resorted documents and materials and additional explanations and videos to the terminal, and the terminal provides them to the user.

[0676] Input: A list of resorted documents and materials and additional explanations and videos.

[0677] Output: The terminal displays the resorted documents and materials and additional explanations and videos to the user.

[0678] (Example 3)

[0679] Next, Example 3 of Form Example 3 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart device 14 is referred to as the "terminal".

[0680] There is a problem that it is difficult for employees to quickly and efficiently search for and organize the necessary documents and materials. In particular, when employees are feeling emotional stress, their search efficiency may further decrease.

[0681] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for analyzing the emotional context of the questions and keywords of employees using an emotion engine, means for providing the search results to the employees, and means for organizing the documents and materials based on the search results. Thereby, employees can quickly and efficiently search for necessary documents and materials and organize information while reducing emotional stress.

[0682] The "natural language processing engine" is software for analyzing the questions and keywords of employees, understanding their meanings, and searching for relevant documents and materials.

[0683] The "emotion engine" is software for analyzing the emotional context contained in the questions and keywords of employees and recognizing their emotional states.

[0684] The "means for searching" is a function of using a natural language processing engine to search for relevant documents and materials from the questions and keywords of employees in a database.

[0685] The "means for providing" is a function of displaying the search results to employees and enabling access to necessary documents and materials.

[0686] The "means for organizing" is a function of categorizing documents and materials based on the search results and automatically storing them in folders or directories corresponding to each category.

[0687] The "machine learning model" is an algorithm that learns based on data and is used when the natural language processing engine analyzes questions and keywords.

[0688] "Categorization" is a process of classifying and organizing documents and materials based on specific criteria.

[0689] "Folders and directories" are digital storage locations for storing documents and materials.

[0690] Modes for Implementing the Invention

[0691] This invention is a system for employees to quickly and efficiently search for and organize necessary documents and materials. The following describes the specific embodiments of this system.

[0692] System Configuration

[0693] This system is composed of three main elements: a server, a terminal, and a user. The server is equipped with a natural language processing engine and an emotion engine, and is responsible for analyzing user input and searching for and organizing relevant documents and materials. The terminal provides an interface for the user to access the system and input questions and keywords.

[0694] Hardware and Software to be Used

[0695] Natural language processing engine: Use natural language processing software such as Google Cloud Natural Language API.

[0696] Emotion engine: Use emotion analysis software such as IBM Watson Tone Analyzer.

[0697] Database: Use a database system for storing documents and materials. Specifically, an SQL database or a NoSQL database can be considered.

[0698] Data Processing and Data Calculation

[0699] The server receives questions and keywords input by the user through the terminal. The received input is first analyzed by the natural language processing engine. This analysis enables understanding the meaning of the input keywords and questions, and searching for relevant documents and materials from the database. At the same time, the sentiment engine analyzes the emotional context contained in the user's input to recognize the user's emotional state.

[0700] Based on the analysis results, the server selects the most relevant documents and materials and categorizes them. The categorized documents and materials are automatically stored in the folders and directories corresponding to their respective categories. For example, documents related to the "Sales Report" category are stored in the "Sales Report" folder.

[0701] Specific Example

[0702] When the user inputs "The latest sales report cannot be found" into the terminal, the processing proceeds as follows.

[0703] 1. The user inputs "The latest sales report cannot be found" into the terminal.

[0704] 2. The terminal sends this input to the server as an HTTP request.

[0705] 3. The server uses the Google Cloud Natural Language API to analyze "The latest sales report" as a keyword.

[0706] 4. The server uses the IBM Watson Tone Analyzer to recognize the user's frustration from the phrase "cannot be found".

[0707] 5. The server searches the database for documents related to the "Sales Report".

[0708] 6. The server automatically stores the search results in the "Sales Report" folder.

[0709] 7. The server generates links to the organized documents and returns them to the terminal.

[0710] 8. The user clicks on the link returned through the terminal and accesses the organized "Business Report".

[0711] Examples of prompt sentences

[0712] "The latest business report cannot be found. Search for relevant documents and organize them in the business report folder."

[0713] With this system, users can quickly and efficiently search for and organize the necessary information. The flow of the specific process in Example 3 will be described with reference to FIG. 21.

[0714] Step 1:

[0715] The user enters a question or keyword into the terminal. For example, enter "The latest business report cannot be found". The input data is sent to the server through the terminal interface.

[0716] Step 2:

[0717] The terminal sends the input to the server. Specifically, an HTTP request is used to send the user's input data to the server. The input data is text data containing the user's question or keyword.

[0718] Step 3:

[0719] The server analyzes the input data using a natural language processing engine. The server analyzes the meaning of the input question or keyword using natural language processing software such as the Google Cloud Natural Language API. As a result of the analysis, relevant keywords and phrases are extracted.

[0720] Step 4:

[0721] The server analyzes the emotional context using an emotion engine. The server analyzes the emotional state included in the user's input using emotion analysis software such as IBM Watson Tone Analyzer. For example, it recognizes the user's frustration from the phrase "not found". As an analysis result, the emotional state is output.

[0722] Step 5:

[0723] The server searches the database for relevant documents and materials. The server searches the database for relevant documents and materials based on the analysis results of the natural language processing engine and the emotion engine. As a search result, a list of relevant documents and materials is output.

[0724] Step 6:

[0725] The server categorizes the search results and stores them in the corresponding folders. The server classifies the search results into specific categories and automatically stores them in the folders or directories corresponding to each category. For example, documents related to the category of "sales report" are stored in the "sales report" folder. A list of the categorized documents and materials is output.

[0726] Step 7:

[0727] The server returns the links to the organized documents and materials to the terminal. The server generates the links to the categorized documents and materials and returns them to the terminal as an HTTP response. The returned links are output in a form that can be accessed by the user.

[0728] Step 8:

[0729] The user accesses the organized documents and materials through the terminal. The user clicks on the link displayed on the terminal and accesses the organized documents and materials. Thereby, the user can obtain the necessary information quickly and efficiently.

[0730] (Application Example 3)

[0731] Next, Application Example 3 of Form Example 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart device 14 is referred to as a "terminal".

[0732] There is a problem that it is difficult for employees to efficiently search for and organize a large amount of documents and materials. In addition, since search results are provided without considering the emotional context of employees, there may be cases where necessary information cannot be quickly found. Especially in an environment such as a logistics center, the management of documents and materials is complicated, and efficient search and organization are required

[0733] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means.

[0734] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to the employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion analysis engine and providing more relevant documents and materials, and means for categorizing the search results and automatically storing them in folders and directories corresponding to each category. As a result, employees can quickly and efficiently search for and organize the necessary information.

[0735] The "natural language processing engine" is a technology for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[0736] The "emotion analysis engine" is a technology for analyzing the emotional context based on the input of employees and providing more relevant documents and materials.

[0737] The "search results" are a set of documents and materials searched by the natural language processing engine.

[0738] "Categorization" is a process of classifying search results into specific categories and automatically storing them in folders or directories corresponding to each category.

[0739] "Folders or directories" are digital storage locations for organizing and storing documents and materials.

[0740] "Employees" refer to users who search for and organize documents and materials using the system.

[0741] "Machine learning model" is an algorithm that learns based on data to improve the performance of natural language processing engines and sentiment analysis engines.

[0742] "Documents and materials" is a general term for texts and digital files containing information related to business.

[0743] The system for implementing this invention includes a natural language processing engine, a sentiment analysis engine, a categorization of search results, and an automatic storage function in folders. Specific embodiments are shown below.

[0744] System Configuration

[0745] The server uses a natural language processing engine to search for relevant documents and materials from employees' questions and keywords. The sentiment analysis engine analyzes the emotional context based on the employees' input and provides more relevant documents and materials. The search results are categorized and automatically stored in folders or directories corresponding to each category.

[0746] Hardware and Software Used

[0747] Hardware: Smartphones, Servers

[0748] Software: Python, Transformers library

[0749] Data processing and data calculation

[0750] The server receives questions and keywords entered by employees using their smartphones. The natural language processing engine analyzes the input text and searches the database for relevant documents and materials. The sentiment analysis engine analyzes the emotional context of the input text to improve the relevance of the search results. The search results are categorized and automatically stored in the corresponding folders or directories.

[0751] Specific example

[0752] When an employee enters "Can't find the latest sales report" using their smartphone, the sentiment analysis engine analyzes the employee's frustration. The natural language processing engine searches for documents related to "sales report", categorizes them, and stores them in a folder. This enables the employee to quickly and efficiently search for and organize the necessary information.

[0753] Example of a prompt sentence

[0754] "Can't find the latest sales report"

[0755] When this prompt sentence is entered, the sentiment analysis engine analyzes the frustration, and the natural language processing engine searches for documents related to "sales report", categorizes them, and stores them in a folder.

[0756] The flow of the specific process in Application Example 3 will be described with reference to FIG. 22.

[0757] Step 1:

[0758] The user enters questions and keywords using their smartphone.

[0759] Input: Questions or keywords entered by the user (e.g., "Can't find the latest sales report")

[0760] Output: Input text data

[0761] Step 2:

[0762] The server receives the input text data and passes it to the natural language processing engine.

[0763] Input: Text data received from the user

[0764] Output: Text data passed to the natural language processing engine

[0765] Step 3:

[0766] The natural language processing engine analyzes the text data and searches the database for relevant documents and materials.

[0767] Input: Text data passed to the natural language processing engine

[0768] Output: List of relevant documents and materials

[0769] Step 4:

[0770] The server passes the text data to the sentiment analysis engine to analyze the emotional context.

[0771] Input: Text data passed to the natural language processing engine

[0772] Output: Sentiment analysis result (e.g., frustration)

[0773] Step 5:

[0774] Based on the results of the sentiment analysis engine, the server selects highly relevant documents and materials.

[0775] Input: List of sentiment analysis results and related documents and materials

[0776] Output: List of highly relevant documents and materials

[0777] Step 6:

[0778] The server categorizes highly relevant documents and materials and automatically stores them in the corresponding folders or directories for each category.

[0779] Input: List of highly relevant documents and materials

[0780] Output: Folders or directories storing categorized documents and materials

[0781] Step 7:

[0782] The server provides the search results to the user and notifies which folder the relevant documents and materials are stored in.

[0783] Input: Information on categorized documents and materials

[0784] Output: Search results and folder information provided to the user

[0785] 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 the voice indicating the user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0786] The data generation model 58 is a so-called generative AI (Artificial Intelligence). Examples of the data generation model 58 include generative AIs such as ChatGPT (registered trademark) (Internet search <URL: https: / / openai.com / blog / chatgpt>). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is input thereto. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization, etc.

[0787] Other examples of generative AI include Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) can be mentioned.

[0788] In the above embodiment, an example of a form in which specific processing is performed by the data processing device 12 is given. However, the technology of the present disclosure is not limited to this, and specific processing may be performed by the smart device 14.

[0789] [Second Embodiment]

[0790] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0791] As shown in FIG. 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.

[0792] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of the "computer" according to the technology of the present disclosure.

[0793] The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. Also, the database 24 and the communication I / F 26 are connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0794] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. Also, the microphone 238, the speaker 240, and the camera 42 are connected to the bus 52.

[0795] The microphone 238 receives instructions and the like from the user 20 by receiving the voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into voice data, and outputs the voice data to the processor 46. The speaker 240 outputs voice according to an instruction from the processor 46.

[0796] The camera 42 is a small digital camera equipped with an optical system such as a lens, an aperture, and a shutter, and an imaging device such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and images the surroundings of the user 20 (for example, an imaging range defined by an angle of view corresponding to the field of view of a general healthy person).

[0797] The communication I / F 44 is connected to the network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0798] FIG. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in FIG. 4, in the data processing device 12, specific processing is performed by the processor 28. The specific processing program 56 is stored in the storage 32.

[0799] The specific processing program 56 is an example of the "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by operating as the specific processing unit 290 according to the specific processing program 56 executed by the processor 28 on the RAM 30.

[0800] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the specific processing unit 290.

[0801] In the smart glasses 214, input / output processing is performed by the processor 46. The storage 50 stores an input / output program 60. The processor 46 reads the input / output program 60 from the storage 50 and executes the read input / output program 60 on the RAM 48. The input / output processing is realized by operating as a control unit 46A according to the input / output program 60 executed by the processor 46 on the RAM 48.

[0802] Next, the specific processing by the specific processing unit 290 of the data processing apparatus 12 will be described.

[0803] "Form Example 1"

[0804] As an embodiment of the present invention, the natural language processing engine analyzes an employee's questions and keywords based on a machine learning model. Specifically, when an employee inputs a keyword such as "latest sales report", the natural language processing engine analyzes this keyword and searches for related documents and materials from the database.

[0805] "Form Example 2"

[0806] The search results are provided to the employee. Specifically, the search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance, so that the employee can quickly find the required information.

[0807] "Form Example 3"

[0808] Furthermore, documents and materials are organized based on the search results. Specifically, the search results are categorized and automatically stored in the folders or directories corresponding to each category. For example, if there is a category of "sales report", the documents and materials related to this category are automatically stored in the "sales report" folder. This enables employees to efficiently search for and organize the necessary information.

[0809] The processing flow of each exemplary form will be described below.

[0810] "Exemplary Form 1"

[0811] Step 1: An employee inputs a question or keywords into the system. For example, keywords such as "latest sales report" are input.

[0812] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. This analysis is performed based on a machine learning model.

[0813] Step 3: The search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance.

[0814] "Exemplary Form 2"

[0815] Step 1: An employee inputs a question or keywords into the system. For example, keywords such as "latest sales report" are input.

[0816] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. This analysis is performed based on a machine learning model.

[0817] Step 3: The search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance.

[0818] Step 4: The selected documents and materials are automatically stored in folders or directories corresponding to their respective categories. For example, if there is a category of "Business Report", the documents and materials related to this category are automatically stored in the "Business Report" folder.

[0819] (Example 1)

[0820] Next, Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[0821] For employees to perform their work efficiently, it is important to quickly search for and obtain the necessary documents and materials. However, in conventional systems, the accuracy of keyword searches is low, and it often takes a long time to find relevant documents and materials. In addition, since the sorting of search results is performed manually, there are problems such as low efficiency and a high likelihood of errors. To solve these problems, a system with a more accurate search function and an automatic sorting function is required.

[0822] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0823] In this invention, the server includes means for searching for relevant documents and materials from an employee's questions or keywords using a natural language processing engine, means for providing the search results to the employee, means for sorting documents and materials based on the search results, means for an employee to input keywords from a terminal, means for the terminal to transmit the input keywords to the server, means for the server to pass the keywords to the natural language processing engine, means for the natural language processing engine to analyze the keywords, means for the server to search a database based on the analysis results, means for the server to return the search results to the terminal, and means for the terminal to display the search results to the employee. As a result, employees can quickly and accurately search for and obtain the necessary documents and materials. In addition, the automatic sorting function of the search results can improve work efficiency and reduce errors.

[0824] The "Natural Language Processing Engine" is software that analyzes human language using a machine learning model to understand its meaning.

[0825] An "employee" is an individual who belongs to a company or organization and conducts business.

[0826] A "question" is the content of an inquiry input by an employee to obtain information.

[0827] A "keyword" is a word or phrase input by an employee to search for specific information.

[0828] A "document" is a material such as a text file or report in which information related to business is described.

[0829] "Materials" are data and documents containing information related to business.

[0830] The "means of searching" is a method for identifying relevant documents and materials using a natural language processing engine.

[0831] The "means of providing" is a method for displaying search results to employees.

[0832] The "means of organizing" is a method of categorizing documents and materials based on search results and storing them in appropriate folders or directories.

[0833] A "terminal" is a device such as a computer or smartphone used by an employee.

[0834] A "server" is a computer system that receives requests from terminals and performs processing.

[0835] A "database" is a collection of information in which documents and materials are stored.

[0836] The "analyzing means" is a method that uses a natural language processing engine to understand the meaning of keywords and extract relevant information.

[0837] The "search results" is a list of relevant documents and materials identified by the natural language processing engine.

[0838] The "displaying means" is a method of visually presenting the search results on a terminal.

[0839] This invention is a system for quickly searching for and obtaining the documents and materials necessary for employees to perform their work efficiently. This system has the function of analyzing the questions and keywords of employees using a natural language processing engine and searching for relevant documents and materials from a database.

[0840] Hardware and software to be used

[0841] Hardware

[0842] Server: A high-performance computer system that manages the database and executes the natural language processing engine.

[0843] Terminal: A device such as a computer or smartphone used by employees.

[0844] Software

[0845] Natural language processing engine: Uses a machine learning model (such as BERT or GPT-3, etc.) to analyze the questions and keywords of employees.

[0846] Database: An aggregate of information where documents and materials are stored, and for example, MySQL, etc. is used.

[0847] Data processing and data calculation

[0848] The server receives the keywords entered by employees from the terminal and sends them to the natural language processing engine. The natural language processing engine analyzes the keywords using a machine learning model and extracts information for identifying relevant documents and materials. The server searches the database based on the analysis results and retrieves relevant documents and materials. The retrieved search results are returned from the server to the terminal and displayed so that employees can view them.

[0849] Specific example

[0850] As a specific example, consider the case where an employee enters "latest sales report". In this case, the server performs the following processing.

[0851] 1. The user enters "latest sales report" in the input field of the terminal.

[0852] 2. The terminal uses an HTTP POST request to send the keywords to the server.

[0853] 3. The server passes the keywords to the natural language processing engine.

[0854] 4. The natural language processing engine analyzes "latest sales report" using the BERT model.

[0855] 5. The server searches the MySQL database based on the analysis results and identifies the latest sales report.

[0856] 6. The server returns the search results to the terminal in JSON format.

[0857] 7. The terminal displays the search results to the user so that the user can view the latest sales report.

[0858] Examples of prompt sentences

[0859] The following are examples of prompt sentences entered by the user.

[0860] "Please display the latest business report."

[0861] "Please provide the sales data for 2023."

[0862] "Please search for the marketing materials of the new product."

[0863] In this way, users can easily obtain the information they need.

[0864] The flow of the specific process in Example 1 will be described with reference to FIG. 11.

[0865] Step 1:

[0866] The user inputs a keyword from the terminal.

[0867] Specifically, the user inputs a keyword such as "the latest business report" into the input field of the terminal. The input keyword is temporarily stored in the memory of the terminal.

[0868] Input: The keyword input by the user (e.g., "the latest business report")

[0869] Output: The keyword stored in the terminal

[0870] Step 2:

[0871] The terminal sends the input keyword to the server.

[0872] Specifically, the terminal uses an HTTP POST request to send the input keyword to the server. At this time, the keyword is included in the request body.

[0873] Input: The keyword stored in the terminal

[0874] Output: The keyword sent to the server

[0875] Step 3:

[0876] The server passes the keyword to the natural language processing engine.

[0877] As a specific operation, the server passes the received keyword to the natural language processing engine. At this time, the keyword is passed as a parameter of the API request.

[0878] Input: Keyword sent to the server

[0879] Output: Keyword passed to the natural language processing engine

[0880] Step 4:

[0881] The natural language processing engine analyzes the keyword.

[0882] As a specific operation, the natural language processing engine uses a machine learning model (e.g., BERT or GPT-3) to analyze the keyword. As an analysis result, information for identifying relevant documents and materials is generated.

[0883] Input: Keyword passed to the natural language processing engine

[0884] Output: Analysis result (information on relevant documents and materials)

[0885] Step 5:

[0886] The server searches the database based on the analysis result.

[0887] As a specific operation, the server searches the database (e.g., MySQL) based on the analysis result. The search query is generated based on the analysis result and executed against the database.

[0888] Input: Analysis result

[0889] Output: Search result (relevant documents and materials) obtained from the database

[0890] Step 6:

[0891] The server returns the search results to the terminal.

[0892] As a specific operation, the server returns the search results obtained from the database to the terminal in a format such as JSON. It is sent as an HTTP response.

[0893] Input: Search results obtained from the database

[0894] Output: Search results sent to the terminal

[0895] Step 7:

[0896] The terminal displays the search results to the user.

[0897] As a specific operation, the terminal displays the search results received from the server on the user interface. For example, links to the latest business reports and previews of the content are displayed.

[0898] Input: Search results sent to the terminal

[0899] Output: Search results displayed to the user (related documents and materials)

[0900] (Application Example 1)

[0901] Next, Application Example 1 of Form 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".

[0902] In a logistics center, there is a problem that it is difficult for employees to quickly obtain necessary information. In particular, there is a lack of means to efficiently search for and provide information such as inventory status, delivery schedules, and the locations of items in the warehouse. As a result, work efficiency decreases, and there is a possibility of work delays and mistakes.

[0903] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0904] In this invention, the server includes means for searching for relevant documents and materials from an employee's questions and keywords using a natural language processing engine, means for providing the search results to the employee, means for sorting the documents and materials based on the search results, means for analyzing the questions input by the employee in voice or text and searching for relevant information from a database, and means for displaying the search results on a smartphone. Thereby, it becomes possible for employees to quickly obtain the necessary information and improve work efficiency.

[0905] The "natural language processing engine" is software for analyzing an employee's questions and keywords and searching for relevant documents and materials.

[0906] The "machine learning model" is an algorithm that learns based on data and analyzes questions and keywords.

[0907] The "database" is a collection of structured data for efficiently searching and obtaining relevant information.

[0908] The "smartphone" is a portable electronic device for analyzing questions input in voice or text and displaying search results.

[0909] The "search result" is relevant information obtained from the database based on the questions and keywords analyzed by the natural language processing engine.

[0910] The "documents and materials" is a collection of texts and data containing information required by employees.

[0911] "Categorization" is a process of classifying documents and materials based on search results and automatically storing them in folders or directories corresponding to each category.

[0912] "The questions entered in voice or text" refer to the content of inquiries entered by employees in voice or text format through a smartphone.

[0913] "Relevant information" refers to the necessary data and materials retrieved from the database for the employees' questions and keywords.

[0914] The system for implementing this invention includes a natural language processing engine, a machine learning model, a database, and a smartphone. The specific configuration and operation of the system will be described below.

[0915] Configuration of the System

[0916] 1. Natural language processing engine: Software for analyzing the questions and keywords entered by employees and retrieving relevant documents and materials. Specifically, natural language processing libraries such as spaCy are used.

[0917] 2. Machine learning model: An algorithm used for analyzing questions and keywords. It is constructed using machine learning libraries such as scikit-learn.

[0918] 3. Database: A collection of structured data for efficiently retrieving and obtaining relevant information. Database management systems such as SQLite are used.

[0919] 4. Smartphone: A portable electronic device for employees to enter questions in voice or text and display search results. iOS or Android smartphones are used.

[0920] Operation of the System

[0921] 1. User input: Employees use a smartphone to enter questions in voice or text format. For example, a prompt sentence such as "Tell me the latest inventory status" is entered.

[0922] 2. Natural Language Processing: The server analyzes the user's input using a natural language processing engine and extracts keywords. For example, keywords such as "inventory status" and "latest" are extracted.

[0923] 3. Database Search: The server searches the database based on the extracted keywords and retrieves relevant documents and materials. For example, a document containing the latest inventory information is searched.

[0924] 4. Providing Search Results: The server sends the search results to the smartphone and displays them to the user. As a result, employees can quickly obtain the necessary information.

[0925] Specific Example

[0926] When an employee enters "Tell me the latest inventory status" into the smartphone, the natural language processing engine extracts keywords such as "inventory status" and "latest", and searches the database for the latest inventory information. As a result, the latest inventory information is displayed on the smartphone.

[0927] Examples of other prompt sentences include questions such as "Tell me the latest delivery schedule" and "Tell me the location of the items in the warehouse".

[0928] With this system, employees can quickly obtain the necessary information and improve work efficiency.

[0929] The flow of the specific process in Application Example 1 will be described with reference to FIG. 12.

[0930] Step 1:

[0931] The user inputs a question in voice or text form using the smartphone.

[0932] Input: User's voice or text form question (e.g., "Tell me the latest inventory status")

[0933] Output: Question data input into the smartphone

[0934] Specific operation: The user launches the smartphone application and performs voice input or text input. In the case of voice input, voice recognition software converts the voice into text.

[0935] Step 2:

[0936] The server analyzes the user's input using a natural language processing engine and extracts keywords.

[0937] Input: User's question data (in text format)

[0938] Output: Extracted keywords (e.g., "inventory status", "latest")

[0939] Specific operation: The server uses a natural language processing library such as spaCy to analyze the text and extract important keywords. For example, it performs morphological analysis to identify important words such as nouns and verbs.

[0940] Step 3:

[0941] The server searches the database based on the extracted keywords and retrieves relevant documents and materials.

[0942] Input: Extracted keywords (e.g., "inventory status", "latest")

[0943] Output: Relevant documents and materials (e.g., documents containing the latest inventory information)

[0944] Specific operation: The server uses a database management system such as SQLite to search for documents and materials that match the keywords. For example, it executes an SQL query to retrieve records that match the keywords.

[0945] Step 4:

[0946] The server sends the search results to the smartphone and displays them to the user.

[0947] Input: Related documents and materials (e.g., documents containing the latest inventory information)

[0948] Output: Search results displayed on the smartphone

[0949] Specific operation: The server converts the search results into a data format such as JSON and sends them to the smartphone. The smartphone application analyzes the received data and displays it in a user-friendly format.

[0950] Step 5:

[0951] The user checks the search results displayed on the smartphone and obtains the necessary information.

[0952] Input: Search results displayed on the smartphone

[0953] Output: Necessary information obtained by the user (e.g., the latest inventory information)

[0954] Specific operation: The user checks the search results displayed on the smartphone screen and obtains the necessary information. For example, check information such as inventory status and delivery schedule.

[0955] (Example 2)

[0956] Next, Example 2 of Embodiment 2 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".

[0957] Employees are required to quickly and efficiently search for necessary documents and materials and provide results sorted in descending order of relevance. However, in conventional systems, there has been a problem that search results are not properly sorted, making it time-consuming to find the required information. In addition, due to the inappropriate display format of search results, there has been an issue that it is difficult for users to quickly find the required information.

[0958] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in the second embodiment is realized by the following respective means.

[0959] In this invention, the server includes means for searching for relevant documents and materials from an employee's question or keyword using a natural language processing engine, means for sorting the search results in descending order of relevance, and means for generating the search results in a list format. As a result, it becomes possible for employees to quickly and efficiently find the required information.

[0960] The "natural language processing engine" is software for analyzing an employee's question or keyword and searching for relevant documents and materials.

[0961] The "searching means" is a function for finding documents and materials in a database based on an employee's question or keyword.

[0962] The "providing means" is a function for displaying search results to an employee.

[0963] The "sorting means" is a function for sorting search results in descending order of relevance.

[0964] The "generating means" is a function for formatting search results in a list format.

[0965] The "displaying means" is a function for displaying the generated search results on an employee's terminal.

[0966] A "machine learning model" is an algorithm that analyzes questions and keywords based on data.

[0967] The "means for categorization" is a function that classifies search results based on specific criteria and automatically stores them in folders or directories corresponding to each category.

[0968] Mode for Implementing the Invention

[0969] This invention is a system that enables employees to quickly and efficiently search for necessary documents and materials and provides results sorted in descending order of relevance. The following describes specific embodiments of this system.

[0970] Configuration of the System

[0971] This system consists of three main elements: a server, a terminal, and a user. The server analyzes search queries using a natural language processing engine and a machine learning model to search for relevant documents and materials. The terminal is a device for the user to input search queries and display search results. The user is an employee who uses the system to search for necessary information.

[0972] Hardware and Software to be Used

[0973] Server: Use a computer system equipped with a high-performance processor and a large-capacity memory. Install Python-based libraries (e.g., NLTK, spaCy) as the natural language processing engine on the server. Also, use Elasticsearch for the search algorithm.

[0974] Terminal: Devices such as personal computers and smartphones used by the user. A web browser or a dedicated application is installed on the terminal.

[0975] Software: A machine learning model (e.g., BERT, GPT) is used to analyze search queries. This enables accurate analysis of user questions and keywords.

[0976] System Operation

[0977] 1. The user inputs a search query.

[0978] The user enters the necessary information in the search bar of the terminal. For example, enter "project report".

[0979] 2. The terminal sends the search query to the server.

[0980] The terminal sends the search query entered by the user to the server. The query is sent using an HTTP request.

[0981] 3. The server receives the search query and queries Elasticsearch.

[0982] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch.

[0983] 4. The server receives search results from Elasticsearch.

[0984] The server receives the search results returned by Elasticsearch. The search results contain information about relevant documents and materials.

[0985] 5. The server sorts the search results in descending order of relevance.

[0986] The server sorts the received search results in descending order of relevance. The scoring function of Elasticsearch is used to evaluate the relevance.

[0987] 6. The server generates the search results in a list format.

[0988] The server formats the sorted search results into a list in HTML format.

[0989] 7. Transmit the search results generated by the server to the terminal

[0990] The server transmits the generated search results in HTML format to the terminal. The results are transmitted using an HTTP response.

[0991] 8. The terminal displays the search results to the user

[0992] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format.

[0993] 9. The user selects the necessary documents and materials

[0994] The user selects the necessary documents and materials from the displayed search results. Click on the selected document to display the details.

[0995] Specific example

[0996] As a specific example, consider the case where the user searches for a "project report". The user enters "project report" from the terminal and clicks the search button. The server receives this query and searches for documents in the database using Elasticsearch. The search results are sorted in descending order of relevance and displayed to the user in a list format. The user can select the necessary documents from this list.

[0997] Examples of prompt sentences

[0998] Examples of prompt sentences may be as follows.

[0999] Prompt sentence: "Please search for a project report"

[1000] When this prompt text is input into the AI model for generating project reports, the AI model provides search results for quickly finding the project reports required by employees.

[1001] The flow of the specific process in Example 2 will be described with reference to FIG. 13.

[1002] Step 1:

[1003] The user inputs a search query.

[1004] The user inputs the required information into the search bar of the terminal. For example, the user inputs "project report". The input query is temporarily stored in the memory of the terminal.

[1005] Step 2:

[1006] The terminal sends the search query to the server.

[1007] The terminal sends the search query input by the user to the server as an HTTP request. This request contains the query input by the user. After sending the request, the terminal waits for a response from the server.

[1008] Step 3:

[1009] The server receives the search query and queries Elasticsearch.

[1010] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch. Specifically, the server passes the search query to the Elasticsearch API and instructs it to search for relevant documents and materials. The input is the search query, and the output is the search results from Elasticsearch.

[1011] Step 4:

[1012] The server receives the search results from Elasticsearch.

[1013] The server receives the search results returned from Elasticsearch. The search results contain information about relevant documents and materials. The server saves this result in memory in preparation for the next process. The input is the search results from Elasticsearch, and the output is the search results saved in the server's memory.

[1014] Step 5:

[1015] The server sorts the search results in descending order of relevance.

[1016] The server sorts the received search results in descending order of relevance using Elasticsearch's scoring function. Specifically, it calculates the relevance score for each document and sorts them in descending order. The input is the search results saved in the server's memory, and the output is the sorted search results.

[1017] Step 6:

[1018] The server generates the search results in a list format.

[1019] The server formats the sorted search results into a list in HTML format. Specifically, it encloses the title and summary of each document with HTML tags to make it user-friendly. The input is the sorted search results, and the output is the search results in HTML format.

[1020] Step 7:

[1021] The server sends the generated search results to the terminal.

[1022] The server sends the generated search results in HTML format to the terminal as an HTTP response. The input is the search results in HTML format, and the output is the search results sent to the terminal.

[1023] Step 8:

[1024] The terminal displays the search results to the user.

[1025] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format. Specifically, the web browser parses the HTML and displays it on the screen. The input is the HTML-formatted search results received from the server, and the output is the search results displayed to the user.

[1026] Step 9:

[1027] The user selects the necessary documents and materials.

[1028] The user selects the necessary documents and materials from the displayed search results. Clicking on the selected document displays the details. The input is the user's click operation, and the output is the detailed display of the selected document.

[1029] (Application Example 2)

[1030] Next, Application Example 2 of Embodiment Example 2 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".

[1031] In a logistics center, it is required that employees quickly search for necessary inventory information and delivery information and efficiently perform their operations. However, in the conventional system, there is a problem that it takes time to search for information, and a large amount of irrelevant information is displayed, resulting in a decrease in work efficiency. Furthermore, since information search using smartphones has not been fully utilized, there is also a problem that it is difficult to respond quickly on-site.

[1032] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1033] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing the documents and materials based on the search results, means for displaying inventory information and delivery information in descending order of relevance, and means for enabling employees to quickly search for necessary information using a smartphone. As a result, employees can quickly search for the necessary information and improve work efficiency.

[1034] The "natural language processing engine" is software for analyzing employees' questions and keywords and searching for relevant documents and materials.

[1035] The "means for providing search results" is a function for displaying the searched documents and materials to employees.

[1036] The "means for organizing documents and materials" is a function for categorizing documents and materials based on the search results and automatically storing them in folders or directories corresponding to each category.

[1037] The "means for displaying inventory information and delivery information in descending order of relevance" is a function for sorting and displaying inventory data and delivery data in descending order of relevance based on a search query.

[1038] The "means for enabling employees to quickly search for necessary information using a smartphone" is a function for enabling employees to quickly search for necessary information using a smartphone.

[1039] The system for implementing this invention enables employees to quickly search for necessary inventory information and delivery information in a logistics center. The following describes specific embodiments of this system.

[1040] Configuration of the System

[1041] This system is composed of the following main components:

[1042] 1. Server: Equipped with a natural language processing engine and a machine learning model, it analyzes employees' questions and keywords.

[1043] 2. Smartphone: A terminal used by employees to input search queries and display search results.

[1044] 3. Database: Stores inventory information and delivery information.

[1045] Program Processing

[1046] The server operates as follows:

[1047] 1. Natural language processing engine: Analyzes questions and keywords input by employees from the smartphone and searches for relevant documents and materials.

[1048] 2. Machine learning model: Generates search results in descending order of relevance based on the analysis of questions and keywords.

[1049] 3. Providing search results: Sends the search results to the smartphone and provides them to employees.

[1050] 4. Data sorting: Categorizes documents and materials based on search results and automatically stores them in the corresponding folders and directories for each category.

[1051] 5. Displaying inventory information and delivery information: Sorts and displays inventory data and delivery data in descending order of relevance.

[1052] Hardware and Software Used

[1053] Hardware: Smartphone, server

[1054] Software: Python, Pandas, Scikit-learn, natural language processing engine (e.g., SpaCy)

[1055] Specific Example

[1056] When an employee searches for "products with insufficient inventory", the system operates as follows:

[1057] 1. The employee enters "products with insufficient inventory" into the smartphone.

[1058] 2. The natural language processing engine of the server analyzes this query and searches for relevant inventory information.

[1059] 3. The machine learning model generates search results in descending order of relevance.

[1060] 4. The search results are displayed on the smartphone, and the employee can view the list of products with insufficient inventory.

[1061] Example of Prompt Sentence

[1062] Search Query: "products with insufficient inventory"

[1063] In this way, employees in the logistics center can quickly search for the necessary information and improve work efficiency.

[1064] The flow of specific processing in Application Example 2 will be described with reference to FIG. 14.

[1065] Step 1:

[1066] The user enters a search query into the smartphone.

[1067] Input: The user enters "products with insufficient inventory" into the smartphone.

[1068] Output: The search query is sent to the server.

[1069] Specific Operation: The user enters "products with insufficient inventory" into the search bar of the smartphone and presses the search button.

[1070] Step 2:

[1071] The server analyzes the search query using a natural language processing engine.

[1072] Input: Search query "out-of-stock products" sent from a smartphone.

[1073] Output: Analyzed query data.

[1074] Specific operation: The server's natural language processing engine (e.g., SpaCy) tokenizes the search query and extracts important keywords.

[1075] Step 3:

[1076] The server searches for relevant documents and materials using a machine learning model.

[1077] Input: Analyzed query data.

[1078] Output: List of relevant documents and materials.

[1079] Specific operation: The server's machine learning model (e.g., Scikit-learn) uses the analyzed query data to search for inventory information and delivery information in the database and sorts them in descending order of relevance.

[1080] Step 4:

[1081] The server sends the search results to the smartphone.

[1082] Input: List of relevant documents and materials.

[1083] Output: Search results displayed on the smartphone.

[1084] Specific operation: The server sends the list of relevant documents and materials to the smartphone and displays them to the user.

[1085] Step 5:

[1086] The user checks the search results on the smartphone.

[1087] Input: Search results displayed on the smartphone.

[1088] Output: The user checks the required information.

[1089] Specific operation: The user scrolls through the search results displayed on the smartphone screen and checks the required inventory information and delivery information.

[1090] Step 6:

[1091] The server sorts out documents and materials based on the search results.

[1092] Input: List of relevant documents and materials.

[1093] Output: Categorized documents and materials.

[1094] Specific operation: The server categorizes documents and materials based on the search results and automatically stores them in the corresponding folders or directories for each category.

[1095] Step 7:

[1096] The server displays the inventory information and delivery information in descending order of relevance.

[1097] Input: Relevant inventory information and delivery information.

[1098] Output: Inventory information and delivery information sorted in descending order of relevance.

[1099] Specific operation: The server sorts the inventory data and delivery data in descending order of relevance and displays it to the user.

[1100] (Example 3)

[1101] Next, Example 3 of Form Example 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[1102] There is a problem that it is difficult for employees to efficiently search for and organize necessary information. In particular, when there are a large number of documents and materials, manual search and organization require time and effort, resulting in a decrease in work efficiency. Also, when the search results are not properly categorized, it is difficult to quickly find the necessary information

[1103] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for searching for relevant information from an employee's question or keyword using a natural language processing engine, means for providing the search results to the employee, means for analyzing and categorizing the information based on the search results, and means for automatically storing the categorized information in folders or directories corresponding to each category. Thereby, employees can efficiently search for and organize necessary information.

[1104] A "natural language processing engine" is software for analyzing natural language and understanding its meaning.

[1105] An "employee" is an individual who belongs to a company or organization and conducts business.

[1106] "Questions and keywords" are input data used by employees when searching for information.

[1107] "Relevant information" refers to documents and materials searched based on an employee's questions and keywords.

[1108] "Means for searching" is a method or technology for obtaining relevant information based on an employee's questions and keywords.

[1109] The "means for providing" refers to the methods and technologies for displaying search results to employees.

[1110] The "means for analyzing and categorizing" refers to the methods and technologies for analyzing search results and classifying them into specific categories.

[1111] "Folders and directories" refer to virtual storage locations for organizing information within a computer.

[1112] The "means for automatically storing" refers to the methods and technologies for automatically moving the analyzed information to the corresponding folders and directories.

[1113] A "machine learning model" is an algorithm that learns based on data and performs predictions and classifications.

[1114] This invention is a system for employees to efficiently search for and organize the necessary information. The following describes the specific embodiments of this system.

[1115] Configuration of the System

[1116] This system consists of three main elements: a server, a terminal, and a user.

[1117] 1. Server:

[1118] The server uses a natural language processing engine to search for relevant information from employees' questions and keywords. Specifically, it uses search engines such as Apache Solr or Elasticsearch. It also uses natural language processing tools such as Google Cloud Natural Language API or IBM Watson Natural Language Understanding to analyze the search results and perform categorization. Furthermore, it uses Python's os module and shutil module to automatically store the categorized information in the corresponding folders and directories.

[1119] 2. Terminal:

[1120] The terminal is a device for the user to input a search query and view the search results. The terminal is composed of computer devices such as personal computers, tablets, and smartphones.

[1121] 3. User:

[1122] The user is an employee who uses the system to search for and organize information. The user inputs a search query using the terminal and views the organized information.

[1123] System Operations

[1124] 1. Input of Search Query by User:

[1125] The user inputs a question or keyword into the search bar of the terminal. For example, input "2023 Business Report".

[1126] 2. Obtaining of Search Results by Server:

[1127] The server receives the search query input by the user and uses Apache Solr or Elasticsearch to obtain relevant information from the database.

[1128] 3. Analysis and Categorization of Search Results by Server:

[1129] The server analyzes the obtained search results using Google Cloud Natural Language API or IBM Watson Natural Language Understanding and classifies them into appropriate categories. For example, classify them into categories such as "Business Report", "Financial Report", "Market Analysis", etc.

[1130] 4. Storage of Each Category into Folders by Server:

[1131] The server automatically stores the categorized information in the folders and directories corresponding to each category using Python's os module and shutil module. For example, the information classified in the "Business Report" category is stored in the "Business Report" folder.

[1132] 5. Confirmation of Organized Information by User:

[1133] The user checks the organized folders using the terminal. For example, the user can open the "Business Report" folder and view the necessary information.

[1134] Examples of Specific Cases and Prompt Sentences

[1135] Specific Case:

[1136] The specific actions when the user searches for the "Business Report for 2023" are as follows.

[1137] 1. User: Enter "Business Report for 2023" in the search bar of the terminal and click the search button.

[1138] 2. Server: Receive the search query and use Apache Solr to obtain relevant information from the database.

[1139] 3. Server: Analyze the obtained information using the Google Cloud Natural Language API and classify it into the "Business Report" category.

[1140] 4. Server: Move the classified information to the "Business Report" folder using Python's os module.

[1141] 5. User: Open the "Business Report" folder on the terminal and confirm the necessary information.

[1142] Example of Prompt Sentence:

[1143] "Search for the 2023 business report and automatically store the relevant information in the business report folder."

[1144] With this system, users can efficiently search for and organize the information they need. The flow of the specific process in Example 3 will be described with reference to FIG. 15.

[1145] Step 1: Input of search query by user

[1146] The user enters a question or keyword in the search bar of the terminal. For example, enter "2023 business report". The input data is the text regarding the information the user wants to search for. The output is sent to the server as a search query.

[1147] Step 2: Obtaining search results by server

[1148] The server receives the search query entered by the user and uses a search engine (e.g., Apache Solr or Elasticsearch) to obtain relevant information from the database. The input data is the user's search query. The server sends the search query to the search engine and obtains relevant documents and materials. The output is a list of documents and materials as search results.

[1149] Step 3: Analysis and categorization of search results by server

[1150] The server analyzes the obtained search results using a natural language processing engine (e.g., Google Cloud Natural Language API or IBM Watson Natural Language Understanding). The input data is the documents and materials obtained as search results. The server analyzes these documents and classifies them into appropriate categories. For example, classify them into categories such as "business report", "financial report", "market analysis", etc. The output is a list of categorized documents and materials.

[1151] Step 4: The server stores the categories in folders

[1152] The server automatically stores categorized documents and materials in folders and directories corresponding to each category. The input data are categorized documents and materials. The server uses Python's os module and shutil module to obtain the document path and move it to the corresponding folder. For example, documents categorized in the "Sales Report" category are stored in the "Sales Report" folder. The output is the documents and materials stored in the folder.

[1153] Step 5: User reviews the organized information

[1154] The user uses a terminal to check the organized folders. The input data are the documents and materials stored in the folders. The user can open, for example, the "Sales Report" folder and view the necessary documents. The output is the documents and materials that the user views.

[1155] This system allows users to efficiently search and organize the information they need.

[1156] (Application example 3)

[1157] Next, a description will be given of Application Example 3 of Form Example 3. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1158] In conventional document management systems, searching and organizing documents and materials is often done manually, resulting in inefficiencies. Furthermore, many paper-based documents are used at logistics centers and other on-site locations, creating a need for a way to digitize and efficiently manage them. Furthermore, categorizing documents and uploading them to cloud storage is often done manually, which is time-consuming and labor-intensive.

[1159] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 3 is realized by the following means.

[1160] In the present invention, the server includes means for searching for relevant documents and materials from an employee's questions and keywords using a natural language processing engine, means for providing the search results to the employee, means for organizing the documents and materials based on the search results, means for extracting text from an image using optical character recognition technology, means for categorizing the documents based on the extracted text and automatically storing them in corresponding folders, and means for uploading the documents to cloud storage. Thereby, efficient search, organization, digitization, and automatic upload of documents and materials to cloud storage become possible.

[1161] The "natural language processing engine" is software for analyzing natural language and searching for relevant information from questions and keywords.

[1162] "Employee" refers to people who belong to a company or organization and perform business operations.

[1163] "Means for searching" refers to a method or technology for finding specific information.

[1164] "Means for providing" refers to a method or technology for displaying the search results to the user or making them accessible.

[1165] "Means for organizing" refers to a method or technology for classifying and managing documents and materials based on specific criteria.

[1166] "Optical character recognition technology" is a technology for extracting character information from an image.

[1167] "Image" refers to a digital file containing visual information.

[1168] "Text" refers to digital data containing character information.

[1169] "Categorization" refers to classifying information based on specific criteria.

[1170] "Folder" refers to a virtual container for organizing digital data.

[1171] "Cloud storage" refers to a remote server for storing data over the Internet.

[1172] "Upload" refers to transferring data from a local device to a remote server.

[1173] The system for implementing this invention is configured as follows. First, the server uses a natural language processing engine to search for relevant documents and materials from employees' questions and keywords. The natural language processing engine analyzes the questions and keywords based on a machine learning model. The analyzed results are provided to the employees.

[1174] Next, the server organizes the documents and materials based on the search results. As a means of organization, optical character recognition technology (OCR) is used to extract text from images. The extracted text is categorized based on specific keywords and automatically stored in the corresponding folders. Furthermore, the documents are uploaded to cloud storage.

[1175] To implement this system, the following hardware and software are used. As hardware, a smartphone with a camera is required. As software, Python, an OCR library, PIL (Python Imaging Library), and Google Cloud Storage are used.

[1176] As a specific example, consider a document sorting app used in a logistics center. When an employee takes a photo of a shipping instruction document with a smartphone camera, the image is converted into text using OCR technology. The converted text is classified into the category of "shipping instruction document" and automatically stored in the corresponding folder. After that, the document is uploaded to cloud storage.

[1177] Examples of prompt sentences may include the following.

[1178] "Please develop a document sorting app used in a logistics center. It is an application that analyzes a document image taken with a smartphone camera using OCR technology, automatically classifies it into categories such as shipping instruction documents, receiving documents, and inventory lists, and has the function of uploading it to cloud storage."

[1179] In this way, efficient search, sorting, digitization of documents and materials, and automatic upload to cloud storage become possible.

[1180] The flow of the specific process in Application Example 3 will be described with reference to FIG. 16.

[1181] Step 1:

[1182] The user takes a photo of a document with the smartphone camera.

[1183] Input: Paper-based document

[1184] Output: Digital image file

[1185] Specific operation: The user launches the smartphone camera app and takes a photo of the document. The captured image is saved in the smartphone.

[1186] Step 2:

[1187] The terminal extracts text from the image using optical character recognition technology (OCR).

[1188] Input: Digital image file

[1189] Output: Extracted text data

[1190] Specific operation: The terminal inputs the saved image file into OCR software and extracts the character information in the image as text data.

[1191] Step 3:

[1192] The terminal analyzes the extracted text and categorizes it.

[1193] Input: Extracted text data

[1194] Output: Category information

[1195] Specific operation: The terminal analyzes the extracted text data and determines the category based on specific keywords (e.g., "shipping instruction", "receipt", etc.).

[1196] Step 4:

[1197] The terminal automatically stores the document in the corresponding folder based on the category.

[1198] Input: Category information, digital image file

[1199] Output: Document stored in the folder

[1200] Specific operation: The terminal moves or copies the digital image file to the corresponding folder based on the determined category.

[1201] Step 5:

[1202] The terminal uploads the document to the cloud storage.

[1203] Input: Document stored in the folder

[1204] Output: Documents saved in cloud storage

[1205] Specific operation: The device uploads the documents in the folder to a cloud storage service (Google Cloud Storage). Once the upload is complete, the documents are stored in the cloud.

[1206] Step 6:

[1207] The server uses a natural language processing engine to search for relevant documents and materials based on the employee's questions and keywords.

[1208] Input: Employee question or keyword

[1209] Output: Search results for related documents and resources

[1210] Specific operation: The server inputs the questions and keywords entered by the employee into a natural language processing engine to search for relevant documents and materials, and provides the search results to the employee.

[1211] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1212] "Example 1"

[1213] In one embodiment of the present invention, a natural language processing engine and an emotion engine are combined. When an employee inputs a question or keyword to the system, the natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. At the same time, the emotion engine analyzes the emotional context of the employee's question or keyword. For example, when an employee inputs the phrase "I can't find the latest sales report", the emotion engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[1214] "Form Example 2"

[1215] In another embodiment of the present invention, the emotion engine sorts documents and materials based on the emotional context of the employee. Specifically, the emotion engine analyzes the emotional context of the employee and automatically stores the documents and materials in the folders or directories corresponding to each emotional context. For example, when an employee inputs the phrase "I can't find the sales report", the emotion engine recognizes the employee's frustration from this phrase, and as a result, the system automatically stores the relevant documents and materials in the "Frustration" folder.

[1216] "Form Example 3"

[1217] In one embodiment of the present invention, a natural language processing engine and an emotion engine are combined. When an employee inputs a question or keyword to the system, the natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database. At the same time, the emotion engine analyzes the emotional context of the employee's question or keyword. For example, when an employee inputs the phrase "I can't find the latest sales report", the emotion engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[1218] The processing flow of each embodiment will be described below.

[1219] "Example 1"

[1220] Step 1: An employee enters a question or keyword into the system.

[1221] Step 2: A natural language processing engine analyzes the keywords and searches the database for relevant documents and materials.

[1222] Step 3: The sentiment engine analyzes the emotional context of the employee's question and keywords.

[1223] Step 4: The natural language processing engine uses the results of the emotion engine's analysis to provide more relevant documents and materials.

[1224] "Example 2"

[1225] Step 1: An employee enters a question or keyword into the system.

[1226] Step 2: The sentiment engine analyzes the emotional context of the employee's question and keywords.

[1227] Step 3: Based on the analysis results of the emotion engine, documents and materials are automatically stored in folders and directories that correspond to their emotional context.

[1228] Example 1

[1229] Next, a description will be given of Example 1 of Form Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1230] Conventional systems were unable to consider emotional context when searching for documents and materials related to questions or keywords entered by employees, making it difficult to fully meet employees' needs. Furthermore, the organization and provision of search results did not provide appropriate information that reflected employees' emotions, hindering efficient work execution.

[1231] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1232] In this invention, the server includes means for searching for relevant documents and materials based on employee questions and keywords using a natural language processing engine, means for providing the search results to the employee, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of the employee questions and keywords using an emotion engine, and means for providing highly relevant documents and materials based on the emotional context. This makes it possible to provide appropriate information that takes employee emotions into consideration, thereby improving work efficiency and employee satisfaction.

[1233] A "natural language processing engine" is software that analyzes questions and keywords entered by employees, understands their meaning, and searches for related documents and materials.

[1234] A "machine learning model" is an algorithm that learns from data and finds patterns and rules, and is used to improve the analytical capabilities of natural language processing engines.

[1235] The "emotion engine" is software that analyzes the emotional context contained in employees' questions and keywords and recognizes those emotions.

[1236] "Documents and materials" is a general term for texts, reports, presentations, data sheets, etc. that contain information employees need to perform their jobs.

[1237] A "database" is an information system for efficiently storing, searching, and managing documents and materials.

[1238] "Search Results" are a list of documents or materials retrieved from a database based on a question or keywords analyzed by a natural language processing engine and sentiment engine.

[1239] "Categorization" is the process of classifying documents or materials based on search results according to specific themes or attributes.

[1240] A "folder or directory" is a virtual storage location on a computer for organizing and storing documents and materials.

[1241] An "employee" is a user who uses the system to input questions or keywords to carry out work.

[1242] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[1243] This invention is a system that analyzes questions and keywords entered by employees and provides relevant documents and materials. The system combines a natural language processing engine and an emotion engine to provide information that takes into account the employee's emotional context.

[1244] Hardware and software used

[1245] server

[1246] The server performs the main processing of the system using the following software:

[1247] Natural language processing engines (e.g., BERT, GPT-3)

[1248] Emotion engine (e.g., sentiment analysis API)

[1249] Database management systems (e.g., MySQL, PostgreSQL)

[1250] Terminal

[1251] The terminal provides an interface for employees to input questions and keywords. The terminal uses the following software:

[1252] Web browser or dedicated application

[1253] HTTP client library

[1254] User

[1255] The user (employee) inputs questions and keywords to the system and checks the search results.

[1256] Data processing and data calculation

[1257] Natural language processing engine

[1258] The server passes the questions and keywords input by the employee to the natural language processing engine for analysis. The natural language processing engine uses a machine learning model (e.g., BERT, GPT-3) to understand the meaning of the input text and extracts keywords for searching relevant documents and materials.

[1259] Sentiment engine

[1260] The server passes the input text to the sentiment engine to analyze the emotional context. The sentiment engine recognizes the employee's emotion (e.g., frustration, joy) from the input text and provides relevant documents and materials based on that information.

[1261] Database search

[1262] The server searches the database for relevant documents and materials based on the analysis results of the natural language processing engine and the sentiment engine. The search results are prioritized considering the employee's emotional context.

[1263] Provision and Arrangement of Results

[1264] The server sends the search results to the terminal, and the terminal displays them to the user. Furthermore, documents and materials are categorized based on the search results and automatically stored in folders or directories corresponding to each category.

[1265] Specific Example

[1266] When an employee inputs "The latest sales report cannot be found", the terminal sends this input to the server. The server analyzes the keyword "the latest sales report" using a natural language processing engine and recognizes the employee's frustration from the phrase "cannot be found" using an emotion engine. Based on this information, the server searches the database for relevant documents and sends the search results to the terminal. The terminal displays the search results to the user.

[1267] Examples of Prompt Sentences

[1268] "Please explain how the system responds when an employee inputs 'The latest sales report cannot be found' to the system."

[1269] By inputting this prompt sentence into the generative AI model, a response that details the operation of the system can be obtained.

[1270] The flow of the specific process in Example 1 will be described with reference to FIG. 17.

[1271] Step 1:

[1272] The user inputs a question or keyword.

[1273] The user enters a question or keyword such as "The latest business report cannot be found" into the system interface. The entered text is displayed in the input field of the terminal. As a specific operation, the user uses the keyboard to enter text and clicks the send button.

[1274] Step 2:

[1275] The terminal sends the input to the server.

[1276] The terminal sends the text entered by the user to the server. The input text is sent to the server as an HTTP request. As a specific operation, the application on the terminal generates an HTTP request and sends it to the server.

[1277] Step 3:

[1278] The server analyzes the input with a natural language processing engine.

[1279] The server passes the received text to the natural language processing engine for analysis. By analyzing the input text, relevant keywords and phrases are extracted. As a specific operation, the server calls the API of the natural language processing engine to obtain the analysis results. The input is the user's text, and the output is the analyzed keywords and phrases.

[1280] Step 4:

[1281] The server analyzes the emotional context with an emotion engine.

[1282] The server passes the input text to the emotion engine to analyze the emotional context. The emotion engine recognizes the emotions of the employees (e.g., frustration) from the input text. As a specific operation, the server calls the API of the emotion engine to obtain the emotion analysis results. The input is the user's text, and the output is the analyzed emotion information.

[1283] Step 5:

[1284] The server searches the database for relevant documents and materials.

[1285] Based on the analysis results of the natural language processing engine and the sentiment engine, the server searches the database for relevant documents and materials. The search results are prioritized considering the emotional context of the employees. As a specific operation, the server generates an SQL query and executes a search against the database. The input is the analysis result, and the output is a list of the retrieved documents and materials.

[1286] Step 6:

[1287] The server sends the search results to the terminal.

[1288] The server sends the search results to the terminal. The search results are sent to the terminal as an HTTP response. As a specific operation, the server generates an HTTP response and sends it to the terminal. The input is the search result, and the output is the data sent to the terminal.

[1289] Step 7:

[1290] The terminal displays the search results to the user.

[1291] The terminal displays the search results received from the server to the user. The user can view the relevant documents and materials. As a specific operation, the application on the terminal displays the search results on the screen. The input is the data from the server, and the output is the search results displayed to the user.

[1292] (Application Example 1)

[1293] Next, Application Example 1 of Form 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".

[1294] It is important for employees to quickly and accurately search for and obtain the necessary documents and materials within the factory. However, in conventional systems, it often took employees a long time to find the information they needed, resulting in reduced efficiency. In addition, since information was provided without considering the emotional state of employees, stress and frustration sometimes increased. To solve these problems, it is necessary to analyze employees' questions and keywords, provide relevant documents and materials, and also provide support considering the emotional context.

[1295] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following respective means.

[1296] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of employees' questions and keywords using an emotion analysis engine, and means for providing additional support to employees based on the emotional context. As a result, employees can not only quickly and accurately obtain the necessary information, but also receive emotional support.

[1297] [[ID=,12]]The "natural language processing engine" is software for analyzing employees' questions and keywords and searching for relevant documents and materials.

[1298] The "machine learning model" is an algorithm that learns based on data and analyzes employees' questions and keywords.

[1299] The "emotion analysis engine" is software for analyzing the emotional context of employees' questions and keywords.

[1300] The "documents and materials" include information such as manuals, work instructions, and reports used within the factory.

[1301] The "search results" refer to a list of relevant documents and materials retrieved by a natural language processing engine.

[1302] "Categorization" is a process of classifying documents and materials into specific categories based on the search results.

[1303] "Folders and directories" are digital storage locations for organizing and storing documents and materials.

[1304] "Additional support" refers to encouraging words and supplementary information provided based on the emotional context analyzed by an emotion analysis engine.

[1305] To implement this invention, the following hardware and software are used. As hardware, a robot body used in the factory, a built-in microphone and speaker, and a display are required. As software, a natural language processing engine (e.g., spaCy, BERT), an emotion analysis engine (e.g., IBM Watson Tone Analyzer), a database management system (e.g., MySQL, PostgreSQL), and robot control software (e.g., ROS - Robot Operating System) are used.

[1306] When an employee inputs a question or keyword to the robot by voice, the built-in microphone captures the voice and converts it into text using speech recognition software (e.g., Google Speech-to-Text API). Next, the natural language processing engine analyzes the converted text and extracts the keyword and the intention of the question. For example, when "Show me the latest work instructions" is input, the keyword "latest work instructions" is extracted.

[1307] After that, the sentiment analysis engine analyzes the emotional context of the text. For example, when "Work instructions cannot be found" is input, frustration is detected. Next, based on the extracted keywords, the server searches the database for relevant documents and materials. For example, it retrieves documents related to "the latest work instructions" from the database.

[1308] The search results are displayed on the display and notified to the employees by voice. Based on the results of the sentiment analysis, encouraging words and additional support are provided as needed. For example, a response such as "You can't find the work instructions and you're having trouble, right? I'll search right away, so please wait a moment." is possible.

[1309] As a specific example, the following examples of prompts are shown.

[1310] Employee: "Show me the latest work instructions."

[1311] Robot: "Searching for the latest work instructions. Please wait a moment."

[1312] (After database search)

[1313] Robot: "Here is the latest work instructions. It is displayed on the screen."

[1314] Also, as a specific example of sentiment analysis, the following conversation can be considered.

[1315] Employee: "I can't find the work instructions. What should I do?"

[1316] Robot: "You can't find the work instructions and you're having trouble, right? I'll search right away, so please wait a moment."

[1317] (After database search)

[1318] Robot: "Here's your latest work order. I've put it up on the screen. Is there anything else I can help you with?"

[1319] In this way, employees not only get the information they need quickly and accurately, but also receive emotional support.

[1320] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1321] Step 1:

[1322] The user inputs questions or keywords to the robot by voice.

[1323] Input: User voice input

[1324] Output: Audio data

[1325] Specific behavior: The user says to the robot, "Show me the latest work instructions."

[1326] Step 2:

[1327] The robot's built-in microphone captures the voice and converts it into text using speech recognition software (e.g., Google Speech-to-Text API).

[1328] Input: Audio data

[1329] Output: Text data

[1330] Specific operation: The robot records the user's voice and converts the voice data into text.

[1331] Step 3:

[1332] The server uses a natural language processing engine (e.g., spaCy, BERT) to analyze the converted text and extract keywords and the intent of the question.

[1333] Input: Text data

[1334] Output: Extracted keywords and intentions

[1335] Specific operation: The server extracts the keyword "latest work instruction" from the text "Show me the latest work instruction".

[1336] Step 4:

[1337] The server uses a sentiment analysis engine (e.g., IBM Watson Tone Analyzer) to analyze the sentiment context of the text.

[1338] Input: Text data

[1339] Output: Sentiment analysis result

[1340] Specific operation: The server detects frustration from the text "The work instruction cannot be found".

[1341] Step 5:

[1342] The server searches the database for relevant documents and materials based on the extracted keywords.

[1343] Input: Extracted keywords

[1344] Output: Search results (relevant documents and materials)

[1345] Specific operation: The server searches the database using the keyword "latest work instruction" and retrieves relevant documents.

[1346] Step 6:

[1347] The server displays the search results on the display and notifies the user audibly.

[1348] Input: Search results

[1349] Output: Display and audio notification

[1350] Specific operation: The server notifies in voice that "This is the latest work instruction. It has been displayed on the screen." and displays the document on the display.

[1351] Step 7:

[1352] Based on the result of sentiment analysis, the server provides encouraging words or additional support as necessary.

[1353] Input: Sentiment analysis result

[1354] Output: Additional support (such as encouraging words)

[1355] Specific operation: The server notifies in voice that "I can't find the work instruction and you seem to be in trouble. I will search immediately, so please wait a moment."

[1356] (Example 2)

[1357] Next, Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[1358] There is a problem that it is difficult for employees to quickly and efficiently find the necessary documents and materials. Also, since information is provided without considering the emotional context of employees, the stress and frustration of employees may increase. Furthermore, since there is a lack of means to organize documents and materials based on the emotions of employees, information management may become complicated.

[1359] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1360] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion analysis engine, and means for automatically storing documents and materials in specific folders or directories based on the emotional context. As a result, employees can quickly and efficiently find the information they need, and by organizing information based on the emotional context, the stress and frustration of employees can be reduced.

[1361] The "natural language processing engine" is software for analyzing the questions and keywords of employees and searching for relevant documents and materials.

[1362] The "machine learning model" is an algorithm that learns patterns based on data and performs predictions and classifications.

[1363] The "emotion analysis engine" is software for analyzing the emotional context from the text input by employees.

[1364] "Folders and directories" are digital storage locations for organizing and storing documents and materials.

[1365] The "search results" are a list of relevant documents and materials returned by the search engine based on the questions and keywords of employees.

[1366] An "employee" is a user who searches for and organizes documents and materials using the system.

[1367] "Documents and materials" are digital files containing information necessary for employees in their work.

[1368] The "emotional context" is the emotion and psychological state that can be read from the text input by employees.

[1369] This invention is a system for employees to quickly and efficiently find the necessary documents and materials. The system consists of a server including a natural language processing engine, a sentiment analysis engine, a search engine, and a database, and a terminal operated by the user.

[1370] The server receives the search query input by the employee from the terminal and analyzes it using the natural language processing engine. The natural language processing engine incorporates a machine learning model to understand the employee's questions and keywords and search for relevant documents and materials. The search engine uses general search software such as Apache Solr or Elasticsearch.

[1371] The search results are displayed in a list format in descending order of relevance. The server provides the search results to the user's terminal, and the user can select the necessary documents and materials.

[1372] Furthermore, the server uses the sentiment analysis engine to analyze the employee's emotional context. The sentiment analysis engine uses sentiment analysis APIs such as IBM Watson or Microsoft Azure. It recognizes the sentiment from the phrases or search queries input by the employee and organizes the documents and materials based on it. For example, when the employee inputs a phrase "Can't find the sales report", the sentiment analysis engine recognizes frustration and automatically stores the relevant documents and materials in the "frustration" folder.

[1373] As a specific example, consider the case where the user inputs "Looking for marketing materials for the new product". The terminal sends this query to the server. The server uses Elasticsearch to search for relevant marketing materials and generates the results in descending order of relevance. At the same time, the server uses the sentiment analysis API of IBM Watson to analyze the sentiment of "expectation". The server displays the search results in a list format on the user's terminal and automatically stores the relevant documents in the "expectation" folder. The user can select and view the necessary materials from the displayed list.

[1374] Examples of prompt sentences are as follows:

[1375] "Looking for marketing materials for new products"

[1376] "Can't find the sales report"

[1377] "Looking for materials summarizing customer feedback"

[1378] In this way, the server can quickly and efficiently provide the information required by the user.

[1379] The flow of the specific process in Example 2 will be described with reference to FIG. 19.

[1380] Step 1:

[1381] The user enters a search query.

[1382] The user enters the required information in the search bar of the terminal. For example, enter "Looking for marketing materials for new products". The entered search query becomes the input data for the next step.

[1383] Step 2:

[1384] The terminal sends the search query to the server.

[1385] The terminal sends the search query entered by the user to the server. An HTTP request is used for this communication. The sent search query becomes the input data for the server.

[1386] Step 3:

[1387] The server generates search results using a search engine.

[1388] The server passes the received search query to a search engine such as Apache Solr or Elasticsearch to search for relevant documents and materials. The search engine returns highly relevant results from the indexed database. The search results become the input data for the next step.

[1389] Step 4:

[1390] The server uses a sentiment analysis engine to analyze the user's sentiment.

[1391] The server uses sentiment analysis APIs such as IBM Watson or Microsoft Azure to analyze the sentiment from the user's search query. For example, it recognizes sentiments such as "expectation" or "excitement" from a query like "Looking for marketing materials for new products". The analyzed sentiment becomes the input data for the next step.

[1392] Step 5:

[1393] The server displays the search results in a list format in descending order of relevance.

[1394] The server sorts the results returned by the search engine in descending order of relevance and displays them on the user's terminal in a list format. The user can select the necessary documents and materials from this list. The displayed search results become the input data for the next step.

[1395] Step 6:

[1396] The server organizes the documents and materials based on sentiment.

[1397] The server automatically stores the relevant documents and materials in specific folders or directories based on the sentiment analyzed by the sentiment analysis engine. For example, if the sentiment of "expectation" is recognized, the relevant documents are stored in the "expectation" folder. The organized documents and materials become the input data for the next step.

[1398] Step 7:

[1399] The user selects the necessary documents and materials.

[1400] The user selects the necessary documents and materials from the displayed list of search results. The selected documents and materials are either downloaded to the user's terminal or displayed in a viewable state. The selected documents and materials become the final output data.

[1401] (Application Example 2)

[1402] Next, Application Example 2 of Morphological Example 2 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[1403] When employees cannot quickly find the necessary documents and materials, there are problems such as a decrease in work efficiency and an increase in stress. In addition, since appropriate support according to the emotional state of employees is not provided, the quality of work may decrease. Especially at the site such as a factory, it is required that the search for work instructions and manuals be performed quickly and accurately, but this is not sufficiently achieved by the current system.

[1404] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following respective means.

[1405] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to the employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion engine and organizing the documents and materials based thereon, and means for sorting relevant documents and materials according to the emotions of employees and providing detailed explanations and videos as necessary. Thereby, employees can quickly find the necessary information, and appropriate support according to the emotional state is provided, so that it is possible to improve work efficiency and reduce stress.

[1406] The "Natural Language Processing Engine" is a technology for analyzing employees' questions and keywords and retrieving relevant documents and materials.

[1407] The "Search Results" are a list of documents and materials retrieved by the natural language processing engine.

[1408] The "Emotion Engine" is a technology for analyzing the emotional context of employees and organizing documents and materials based on it.

[1409] The "Emotional Context" is the emotional state inferred from the questions and keywords input by employees.

[1410] "Sorting" means arranging documents and materials based on specific criteria.

[1411] "Detailed Explanation" means providing additional information and specific procedures for documents and materials.

[1412] "Video" is video content for providing information visually.

[1413] "Folder" is a virtual storage location for organizing documents and materials.

[1414] "Directory" is a structure for organizing files and folders in a computer system.

[1415] The system for implementing this invention has the following configuration. The server includes a natural language processing engine, an emotion engine, a document search engine, and a database. The terminal includes a voice input device, a display, and an interface. The user accesses the system through the terminal and inputs questions and keywords.

[1416] The server first uses a natural language processing engine to analyze the question and keywords entered by the user. Based on this analysis, a document search engine searches the database for relevant documents and materials. The search results are sorted in order of relevance and provided to the user.

[1417] The emotion engine then analyzes the emotional context from the user's input. For example, if the user types, "I don't know how to install this part," the emotion engine recognizes the user's frustration. Based on this emotional context, the server further sorts relevant documents and resources, providing detailed instructions or videos as needed.

[1418] For example, consider a factory worker typing, "I don't know how to install this part." In this case, the server searches for relevant manuals, and if the emotion engine recognizes the worker's frustration, it provides detailed instructions and videos. This allows the worker to quickly obtain the information they need, improving work efficiency.

[1419] An example of a prompt to input to a generative AI model is as follows:

[1420] Create a program that analyzes the sentiment of employees when they type the phrase "I don't know how to install this part" and searches for and displays relevant documents and manuals. If employees are frustrated, provide detailed instructions or videos as well.

[1421] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1422] Step 1:

[1423] The user inputs a question or keyword through the terminal.

[1424] Input: The user enters a question or keyword by voice or text.

[1425] Output: The terminal sends the input questions and keywords to the server.

[1426] Step 2:

[1427] The server uses a natural language processing engine to analyze the user's questions and keywords.

[1428] Input: Questions and keywords sent from the terminal.

[1429] Data processing: The natural language processing engine uses a machine learning model to analyze the meaning of the questions and keywords.

[1430] Output: Keywords and phrases as analysis results.

[1431] Step 3:

[1432] The server uses a document search engine to search the database for relevant documents and materials based on the analysis results.

[1433] Input: Analysis results of the natural language processing engine.

[1434] Data calculation: The document search engine searches for documents and materials in the database and lists them in descending order of relevance.

[1435] Output: List of relevant documents and materials.

[1436] Step 4:

[1437] The server sends the search results to the terminal, and the terminal provides them to the user.

[1438] Input: List of relevant documents and materials.

[1439] Output: The terminal displays the search results to the user.

[1440] Step 5:

[1441] The server uses an emotion engine to analyze the emotional context from the user's input.

[1442] Input: The user's question or keyword.

[1443] Data processing: The emotion engine analyzes the emotion from the user's input and identifies the emotional context.

[1444] Output: Emotional context (e.g., frustration).

[1445] Step 6:

[1446] The server resorts the relevant documents and materials based on the emotional context and adds detailed explanations and videos if necessary.

[1447] Input: The list of documents and materials related to the emotional context.

[1448] Data calculation: Resort the documents and materials based on the emotional context and add detailed explanations and videos.

[1449] Output: The list of resorted documents and materials and additional explanations and videos.

[1450] Step 7:

[1451] The server sends the list of resorted documents and materials and additional explanations and videos to the terminal, and the terminal provides them to the user.

[1452] Input: The list of resorted documents and materials and additional explanations and videos.

[1453] Output: The terminal displays the resorted documents and materials and additional explanations and videos to the user.

[1454] (Example 3)

[1455] Next, Example 3 of Embodiment 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[1456] There is a problem that it is difficult for employees to quickly and efficiently search for and organize necessary documents and materials. In particular, when employees are feeling emotional stress, the search efficiency may further decrease.

[1457] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for searching for relevant documents and materials from an employee's question or keyword using a natural language processing engine, means for analyzing the emotional context of an employee's question or keyword using an emotion engine, means for providing the search results to the employee, and means for organizing documents and materials based on the search results. Thereby, employees can quickly and efficiently search for necessary documents and materials and organize information while reducing emotional stress.

[1458] The "natural language processing engine" is software for analyzing an employee's question or keyword, understanding its meaning, and searching for relevant documents and materials.

[1459] The "emotion engine" is software for analyzing the emotional context included in an employee's question or keyword and recognizing their emotional state.

[1460] The "means for searching" is a function of using a natural language processing engine to find relevant documents and materials from an employee's question or keyword in a database.

[1461] The "means for providing" is a function of displaying the search results to the employee and enabling access to necessary documents and materials.

[1462] The "sorting means" is a function that categorizes documents and materials based on search results and automatically stores them in folders or directories corresponding to each category.

[1463] The "machine learning model" is an algorithm that learns based on data and is used when the natural language processing engine analyzes questions and keywords.

[1464] "Categorization" is a process of classifying and organizing documents and materials based on specific criteria.

[1465] "Folders and directories" are digital storage locations for storing documents and materials.

[1466] Modes for Implementing the Invention

[1467] This invention is a system for employees to quickly and efficiently search for and organize necessary documents and materials. The following describes specific embodiments of this system.

[1468] Configuration of the System

[1469] This system is composed of three main elements: a server, a terminal, and a user. The server is equipped with a natural language processing engine and an emotion engine, and is responsible for analyzing user input and searching for and organizing relevant documents and materials. The terminal provides an interface for the user to access the system and input questions and keywords.

[1470] Hardware and Software to be Used

[1471] Natural language processing engine: Use natural language processing software such as Google Cloud Natural Language API.

[1472] Emotion engine: Use emotion analysis software such as IBM Watson Tone Analyzer.

[1473] Database: Use a database system for storing documents and materials. Specifically, an SQL database or a NoSQL database can be considered.

[1474] Data processing and data calculation

[1475] The server receives questions and keywords input by the user through the terminal. The received input is first analyzed by a natural language processing engine. Through this analysis, the meaning of the input keywords and questions is understood, and relevant documents and materials are retrieved from the database. At the same time, the sentiment engine analyzes the emotional context contained in the user's input to recognize the user's emotional state.

[1476] Based on the analysis results, the server selects the most relevant documents and materials and categorizes them. The categorized documents and materials are automatically stored in the folders or directories corresponding to their respective categories. For example, documents related to the "Sales Report" category are stored in the "Sales Report" folder.

[1477] Specific example

[1478] When the user inputs "Can't find the latest sales report" on the terminal, the following processing proceeds.

[1479] 1. The user inputs "Can't find the latest sales report" on the terminal.

[1480] 2. The terminal sends this input to the server as an HTTP request.

[1481] 3. The server uses the Google Cloud Natural Language API to analyze "the latest sales report" as a keyword.

[1482] 4. The server uses IBM Watson Tone Analyzer to recognize the user's frustration from the phrase "not found".

[1483] 5. The server searches the database for documents related to the "sales report".

[1484] 6. The server automatically stores the search results in the "sales report" folder.

[1485] 7. The server generates a link to the organized document and returns it to the terminal.

[1486] 8. The user clicks on the link returned through the terminal and accesses the organized "sales report".

[1487] Example of a prompt sentence

[1488] "The latest sales report cannot be found. Please search for related documents and organize them in the sales report folder."

[1489] With this system, the user can search and organize the necessary information quickly and efficiently. The flow of the specific process in Example 3 will be described with reference to Figure 21.

[1490] Step 1:

[1491] The user enters a question or keyword into the terminal. For example, enter "The latest sales report cannot be found". The input data is sent to the server through the terminal interface.

[1492] Step 2:

[1493] The terminal sends the input to the server. Specifically, the user's input data is sent to the server using an HTTP request. The input data is text data containing the user's question or keyword.

[1494] Step 3:

[1495] The server analyzes the input data using a natural language processing engine. The server uses natural language processing software such as the Google Cloud Natural Language API to analyze the meaning of the input questions or keywords. As a result of the analysis, relevant keywords and phrases are extracted.

[1496] Step 4:

[1497] The server analyzes the emotional context using an emotion engine. The server uses emotion analysis software such as the IBM Watson Tone Analyzer to analyze the emotional state included in the user's input. For example, the frustration of the user is recognized from the phrase "not found". As a result of the analysis, the emotional state is output.

[1498] Step 5:

[1499] The server searches the database for relevant documents and materials. The server searches the database for relevant documents and materials based on the analysis results of the natural language processing engine and the emotion engine. As a result of the search, a list of relevant documents and materials is output.

[1500] Step 6:

[1501] The server categorizes the search results and stores them in the corresponding folders. The server classifies the search results into specific categories and automatically stores them in the folders or directories corresponding to each category. For example, documents related to the category of "sales report" are stored in the "sales report" folder. A list of the categorized documents and materials is output.

[1502] Step 7:

[1503] The server returns to the terminal the links to the organized documents and materials. The server generates the links to the categorized documents and materials and returns them to the terminal as an HTTP response. The returned links are output in a form accessible by the user.

[1504] Step 8:

[1505] The user accesses the organized documents and materials through the terminal. The user clicks on the link displayed on the terminal to access the organized documents and materials. Thereby, the user can obtain the necessary information quickly and efficiently.

[1506] (Application Example 3)

[1507] Next, Application Example 3 of Embodiment Example 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the smart glasses 214 are referred to as a "terminal".

[1508] There is a problem that it is difficult for employees to efficiently search for and organize a large amount of documents and materials. Also, since search results are provided without considering the emotional context of employees, there are cases where the necessary information cannot be found quickly. In particular, in an environment such as a logistics center, the management of documents and materials is complicated, and efficient search and organization are required.

[1509] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following respective means.

[1510] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of employees using an emotion analysis engine and providing more relevant documents and materials, and means for categorizing the search results and automatically storing them in folders or directories corresponding to each category. Thereby, employees can search for and organize the necessary information quickly and efficiently.

[1511] The "natural language processing engine" is a technology for analyzing employees' questions and keywords and searching for relevant documents and materials.

[1512] The "emotion analysis engine" is a technology for analyzing the emotional context based on employees' inputs and providing more relevant documents and materials.

[1513] The "search results" refer to the collection of documents and materials searched by the natural language processing engine.

[1514] "Categorization" is a process of classifying the search results into specific categories and automatically storing them in folders or directories corresponding to each category.

[1515] "Folders or directories" are digital storage locations for organizing and storing documents and materials.

[1516] "Employees" refer to users who search for and organize documents and materials using the system.

[1517] The "machine learning model" is an algorithm that learns based on data to improve the performance of the natural language processing engine and the emotion analysis engine.

[1518] "Documents and materials" is a general term for texts and digital files containing information related to business.

[1519] The system for implementing this invention includes a natural language processing engine, a sentiment analysis engine, and a function for categorizing search results and automatically storing them in folders. Specific embodiments are shown below.

[1520] System Configuration

[1521] The server uses a natural language processing engine to search for relevant documents and materials from employees' questions and keywords. The sentiment analysis engine analyzes the emotional context based on the employees' input and provides more relevant documents and materials. The search results are categorized and automatically stored in the corresponding folders and directories.

[1522] Hardware and Software Used

[1523] Hardware: Smartphones, Servers

[1524] Software: Python, Transformers Library

[1525] Data Processing and Data Calculation

[1526] The server receives questions and keywords input by employees using smartphones. The natural language processing engine analyzes the input text and searches for relevant documents and materials from the database. The sentiment analysis engine analyzes the emotional context of the input text and improves the relevance of the search results. The search results are categorized and automatically stored in the corresponding folders and directories.

[1527] Specific Example

[1528] When an employee uses a smartphone to enter "The latest business report cannot be found", the sentiment analysis engine analyzes the employee's frustration. The natural language processing engine searches for documents related to "business report", categorizes them, and stores them in folders. This enables employees to quickly and efficiently search for and organize the necessary information.

[1529] Examples of prompt sentences

[1530] "The latest business report cannot be found"

[1531] When this prompt sentence is entered, the sentiment analysis engine analyzes the frustration, and the natural language processing engine searches for documents related to "business report", categorizes them, and stores them in folders.

[1532] The flow of specific processing in Application Example 3 will be described with reference to FIG. 22.

[1533] Step 1:

[1534] The user uses a smartphone to enter questions or keywords.

[1535] Input: Questions or keywords entered by the user (e.g., "The latest business report cannot be found")

[1536] Output: Input text data

[1537] Step 2:

[1538] The server receives the input text data and passes it to the natural language processing engine.

[1539] Input: Text data received from the user

[1540] Output: Text data passed to the natural language processing engine

[1541] Step 3:

[1542] The natural language processing engine analyzes the text data and searches for relevant documents and materials from the database.

[1543] Input: Text data passed to the natural language processing engine

[1544] Output: List of relevant documents and materials

[1545] Step 4:

[1546] The server passes the text data to the sentiment analysis engine to analyze the emotional context.

[1547] Input: Text data passed to the natural language processing engine

[1548] Output: Sentiment analysis result (e.g., frustration)

[1549] Step 5:

[1550] Based on the results of the sentiment analysis engine, the server selects highly relevant documents and materials.

[1551] Input: Sentiment analysis result and list of relevant documents and materials

[1552] Output: List of highly relevant documents and materials

[1553] Step 6:

[1554] The server categorizes the highly relevant documents and materials and automatically stores them in the corresponding folders or directories for each category.

[1555] Input: List of highly relevant documents and materials

[1556] Output: Folders or directories storing the categorized documents and materials

[1557] Step 7:

[1558] The server provides the user with search results and notifies the user of the folder where the relevant documents and materials are stored.

[1559] Input: Information of categorized documents and materials

[1560] Output: Search results and folder information provided to the user

[1561] 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 voice indicating user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1562] The data generation model 58 is a so-called generative AI (Artificial Intelligence). The Examples of the data generation model 58 include generative AIs such as ChatGPT (Internet search <URL: https: / / openai.com / blog / chatgpt>). The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including instructions is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is input. The data generation model 58 infers the input inference data according to the instructions indicated by the prompt and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization, etc.

[1563] Other examples of generative AIs include Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) can be mentioned.

[1564] In the above embodiment, an example of a form in which specific processing is performed by the data processing device 12 is given. However, the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1565] [Third Embodiment]

[1566] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1567] As shown in FIG. 5, the data processing system 310 includes a data processing device 12 and a headset-type terminal 314. An example of the data processing device 12 is a server.

[1568] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of the "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. Also, the database 24 and the communication I / F 26 are connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1569] The headset-type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. Also, the microphone 238, the speaker 240, the camera 42, and the display 343 are connected to the bus 52.

[1570] The microphone 238 receives instructions and the like from the user 20 by receiving the voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into voice data, and outputs the voice data to the processor 46. The speaker 240 outputs voice according to an instruction from the processor 46.

[1571] The camera 42 is a small digital camera equipped with an optical system such as a lens, an aperture, and a shutter, and an imaging device such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and images the surroundings of the user 20 (for example, an imaging range defined by an angle of view corresponding to the field of view of a general healthy person).

[1572] The communication I / F 44 is connected to the network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[1573] FIG. 6 shows an example of the main functions of the data processing device 12 and the headset-type terminal 314. As shown in FIG. 6, in the data processing device 12, specific processing is performed by the processor 28. The specific processing program 56 is stored in the storage 32.

[1574] The specific processing program 56 is an example of the "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by operating as the specific processing unit 290 according to the specific processing program 56 executed by the processor 28 on the RAM 30.

[1575] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the specific processing unit 290.

[1576] In the headset type terminal 314, input / output processing is performed by the processor 46. The storage 50 stores an input / output program 60. The processor 46 reads out the input / output program 60 from the storage 50 and executes the read input / output program 60 on the RAM 48. The input / output processing is realized by operating as a control unit 46A according to the input / output program 60 that the processor 46 executes on the RAM 48.

[1577] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1578] "Form Example 1"

[1579] As an embodiment of the present invention, the natural language processing engine analyzes an employee's question or keyword based on a machine learning model. Specifically, when an employee inputs a keyword such as "latest sales report", the natural language processing engine analyzes this keyword and searches for related documents and materials from the database.

[1580] "Form Example 2"

[1581] The search results are provided to the employee. Specifically, the search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance, and the employee can quickly find the required information.

[1582] "Form Example 3"

[1583] Furthermore, documents and materials are organized based on the search results. Specifically, the search results are categorized and automatically stored in the corresponding folders or directories. For example, if there is a category of "sales report", the documents and materials related to this category are automatically stored in the "sales report" folder. This enables employees to efficiently search for and organize the necessary information.

[1584] The processing flow of each exemplary embodiment will be described below.

[1585] "Exemplary Embodiment 1"

[1586] Step 1: An employee inputs a question or keyword to the system. For example, a keyword such as "latest sales report" is input.

[1587] Step 2: The natural language processing engine analyzes this keyword and searches for related documents and materials from the database. This analysis is performed based on a machine learning model.

[1588] Step 3: The search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance.

[1589] "Exemplary Embodiment 2"

[1590] Step 1: An employee inputs a question or keyword to the system. For example, a keyword such as "latest sales report" is input.

[1591] Step 2: The natural language processing engine analyzes this keyword and searches for related documents and materials from the database. This analysis is performed based on a machine learning model.

[1592] Step 3: The search results are displayed in a list format, and the employee can select the necessary documents and materials. Also, the search results are displayed in descending order of relevance.

[1593] Step 4: The selected documents and materials are automatically stored in the folders or directories corresponding to their respective categories. For example, if there is a category of "Business Report", the documents and materials related to this category are automatically stored in the "Business Report" folder.

[1594] (Example 1)

[1595] Next, Example 1 of Form Example 1 will be described. In the following description, the data processing device 12 is referred to as a "server", and the headset type terminal 314 is referred to as a "terminal".

[1596] For employees to perform their work efficiently, it is important to quickly search for and obtain the necessary documents and materials. However, in conventional systems, the accuracy of keyword searches is low, and it often takes a long time to find relevant documents and materials. In addition, since the sorting of search results is performed manually, there are problems such as low efficiency and a high likelihood of errors. To solve these problems, a system with a more accurate search function and an automatic sorting function is required.

[1597] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1598] In this invention, the server includes means for searching for relevant documents and materials from an employee's questions or keywords using a natural language processing engine, means for providing the search results to the employee, means for sorting documents and materials based on the search results, means for an employee to input keywords from a terminal, means for the terminal to send the input keywords to the server, means for the server to pass the keywords to the natural language processing engine, means for the natural language processing engine to analyze the keywords, means for the server to search a database based on the analysis results, means for the server to return the search results to the terminal, and means for the terminal to display the search results to the employee. As a result, employees can quickly and accurately search for and obtain the necessary documents and materials. In addition, the automatic sorting function of search results can improve work efficiency and reduce errors.

[1599] The "natural language processing engine" is software that analyzes human language using a machine learning model to understand its meaning.

[1600] An "employee" is an individual who belongs to a company or organization and conducts business.

[1601] A "question" is the content of an inquiry input by an employee to obtain information.

[1602] A "keyword" is a word or phrase input by an employee to search for specific information.

[1603] A "document" is a material such as a text file or report in which information related to business is described.

[1604] "Materials" are data and documents containing information related to business.

[1605] The "means for searching" is a method for identifying relevant documents and materials using a natural language processing engine.

[1606] The "means for providing" is a method for displaying search results to employees.

[1607] The "means for organizing" is a method for categorizing documents and materials based on search results and storing them in appropriate folders or directories.

[1608] A "terminal" is a device such as a computer or smartphone used by an employee.

[1609] A "server" is a computer system that receives requests from terminals and performs processing.

[1610] A "database" is an aggregate of information in which documents and materials are stored.

[1611] The "means for analysis" is a method of understanding the meaning of keywords using a natural language processing engine and extracting relevant information.

[1612] The "search results" are a list of relevant documents and materials identified by a natural language processing engine.

[1613] The "means for display" is a method of visually presenting search results on a terminal.

[1614] This invention is a system for quickly searching for and obtaining documents and materials necessary for employees to efficiently perform their work. This system has the function of analyzing employees' questions and keywords using a natural language processing engine and searching for relevant documents and materials from a database.

[1615] Hardware and software to be used

[1616] Hardware

[1617] Server: A high-performance computer system that manages the database and executes the natural language processing engine.

[1618] Terminal: A device such as a computer or smartphone used by employees.

[1619] Software

[1620] Natural language processing engine: Uses machine learning models (such as BERT or GPT-3, etc.) to analyze employees' questions and keywords.

[1621] Database: An aggregate of information storing documents and materials, and for example, MySQL, etc. is used.

[1622] Data processing and data calculation

[1623] The server receives the keywords entered by employees from the terminal and sends them to the natural language processing engine. The natural language processing engine analyzes the keywords using a machine learning model and extracts information for identifying relevant documents and materials. The server searches the database based on the analysis results and retrieves the relevant documents and materials. The retrieved search results are returned from the server to the terminal and displayed so that employees can view them.

[1624] Specific example

[1625] As a specific example, consider the case where an employee enters "latest sales report". In this case, the server performs the following processing.

[1626] 1. The user enters "latest sales report" in the input field of the terminal.

[1627] 2. The terminal uses an HTTP POST request to send the keywords to the server.

[1628] 3. The server passes the keywords to the natural language processing engine.

[1629] 4. The natural language processing engine analyzes "latest sales report" using the BERT model.

[1630] 5. The server searches the MySQL database based on the analysis results and identifies the latest sales report.

[1631] 6. The server returns the search results to the terminal in JSON format.

[1632] 7. The terminal displays the search results to the user so that the user can view the latest sales report.

[1633] Examples of prompt sentences

[1634] The following are examples of prompt sentences entered by the user.

[1635] "Please display the latest business report."

[1636] "Please provide the sales data for 2023."

[1637] "Please search for the marketing materials of the new product."

[1638] In this way, users can easily obtain the information they need.

[1639] The flow of the specific process in Example 1 will be described with reference to FIG. 11.

[1640] Step 1:

[1641] The user inputs a keyword from the terminal.

[1642] Specifically, the user inputs a keyword such as "the latest business report" into the input field of the terminal. The input keyword is temporarily stored in the memory of the terminal.

[1643] Input: The keyword input by the user (e.g., "the latest business report")

[1644] Output: The keyword stored in the terminal

[1645] Step 2:

[1646] The terminal sends the input keyword to the server.

[1647] Specifically, the terminal uses an HTTP POST request to send the input keyword to the server. At this time, the keyword is included in the request body.

[1648] Input: The keyword stored in the terminal

[1649] Output: The keyword sent to the server

[1650] Step 3:

[1651] The server passes the keyword to the natural language processing engine.

[1652] Specifically, the server passes the received keyword to the natural language processing engine. At this time, the keyword is passed as a parameter of the API request.

[1653] Input: Keyword sent to the server

[1654] Output: Keyword passed to the natural language processing engine

[1655] Step 4:

[1656] The natural language processing engine analyzes the keyword.

[1657] Specifically, the natural language processing engine uses a machine learning model (e.g., BERT or GPT-3) to analyze the keyword. As an analysis result, information for identifying relevant documents and materials is generated.

[1658] Input: Keyword passed to the natural language processing engine

[1659] Output: Analysis result (information on relevant documents and materials)

[1660] Step 5:

[1661] The server searches the database based on the analysis result.

[1662] Specifically, the server searches the database (e.g., MySQL) based on the analysis result. The search query is generated based on the analysis result and executed against the database.

[1663] Input: Analysis result

[1664] Output: Search result (relevant documents and materials) obtained from the database

[1665] Step 6:

[1666] The server returns the search results to the terminal.

[1667] Specifically, the server returns the search results obtained from the database to the terminal in a format such as JSON. It is sent as an HTTP response.

[1668] Input: Search results obtained from the database

[1669] Output: Search results sent to the terminal

[1670] Step 7:

[1671] The terminal displays the search results to the user.

[1672] Specifically, the terminal displays the search results received from the server on the user interface. For example, links to the latest business reports and previews of the content are displayed.

[1673] Input: Search results sent to the terminal

[1674] Output: Search results displayed to the user (related documents and materials)

[1675] (Application Example 1)

[1676] Next, Application Example 1 of Embodiment 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset-type terminal 314 is referred to as the "terminal".

[1677] In a logistics center, there is a problem that it is difficult for employees to quickly obtain the necessary information. In particular, there is a lack of means to efficiently search for and provide information such as inventory status, delivery schedules, and the locations of items in the warehouse. As a result, work efficiency decreases, and there is a possibility of work delays and mistakes.

[1678] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following respective means.

[1679] In this invention, the server includes means for searching for relevant documents and materials from an employee's questions and keywords using a natural language processing engine, means for providing the search results to the employee, means for organizing the documents and materials based on the search results, means for analyzing questions input by the employee in voice or text and searching for relevant information from a database, and means for displaying the search results on a smartphone. Thereby, it becomes possible for the employee to quickly obtain necessary information and improve work efficiency.

[1680] The "natural language processing engine" is software for analyzing an employee's questions and keywords and searching for relevant documents and materials.

[1681] The "machine learning model" is an algorithm that learns based on data and analyzes questions and keywords.

[1682] The "database" is a collection of structured data for efficiently searching and obtaining relevant information.

[1683] The "smartphone" is a portable electronic device for analyzing questions input in voice or text and displaying search results.

[1684] The "search result" is relevant information obtained from the database based on questions and keywords analyzed by the natural language processing engine.

[1685] The "documents and materials" is a collection of texts and data including information required by the employee.

[1686] "Categorization" is a process of classifying documents and materials based on search results and automatically storing them in folders or directories corresponding to each category.

[1687] The "questions input in voice or text" refer to the content of inquiries input by employees in voice or text form through smartphones.

[1688] The "related information" refers to the necessary data and materials retrieved from the database for the employees' questions and keywords.

[1689] The system for implementing this invention includes a natural language processing engine, a machine learning model, a database, and a smartphone. The specific configuration and operation of the system will be described below.

[1690] Configuration of the System

[1691] 1. Natural language processing engine: Software for analyzing the questions and keywords input by employees and retrieving relevant documents and materials. Specifically, natural language processing libraries such as spaCy are used.

[1692] 2. Machine learning model: An algorithm used for analyzing questions and keywords. It is constructed using machine learning libraries such as scikit-learn.

[1693] 3. Database: A collection of structured data for efficiently searching and obtaining relevant information. Database management systems such as SQLite are used.

[1694] 4. Smartphone: A portable electronic device for employees to input questions in voice or text and display search results. iOS or Android smartphones are used.

[1695] Operation of the System

[1696] 1. User input: Employees input questions in voice or text form using a smartphone. For example, a prompt sentence such as "Tell me the latest inventory status" is input.

[1697] 2. Natural Language Processing: The server analyzes the user's input using a natural language processing engine and extracts keywords. For example, keywords such as "inventory status" and "latest" are extracted.

[1698] 3. Database Search: The server searches the database based on the extracted keywords and retrieves relevant documents and materials. For example, a document containing the latest inventory information is searched.

[1699] 4. Providing Search Results: The server sends the search results to the smartphone and displays them to the user. As a result, employees can quickly obtain the necessary information.

[1700] Specific Example

[1701] When an employee enters "Tell me the latest inventory status" into the smartphone, the natural language processing engine extracts keywords such as "inventory status" and "latest", and searches the database for the latest inventory information. As a result, the latest inventory information is displayed on the smartphone.

[1702] As examples of other prompt sentences, questions such as "Tell me the latest delivery schedule" and "Tell me the location of the items in the warehouse" can be considered.

[1703] With this system, employees can quickly obtain the necessary information and improve work efficiency.

[1704] The flow of the specific process in Application Example 1 will be described with reference to FIG. 12.

[1705] Step 1:

[1706] The user inputs a question in voice or text form using the smartphone.

[1707] Input: User's voice or text-form question (e.g., "Tell me the latest inventory status")

[1708] Output: Question data input into the smartphone

[1709] Specific operation: The user launches the application on the smartphone and performs voice input or text input. In the case of voice input, voice recognition software converts the voice into text.

[1710] Step 2:

[1711] The server analyzes the user's input using a natural language processing engine and extracts keywords.

[1712] Input: User's question data (in text format)

[1713] Output: Extracted keywords (e.g., "inventory status", "latest")

[1714] Specific operation: The server uses a natural language processing library such as spaCy to analyze the text and extract important keywords. For example, it performs morphological analysis to identify important words such as nouns and verbs.

[1715] Step 3:

[1716] The server searches the database based on the extracted keywords and retrieves relevant documents and materials.

[1717] Input: Extracted keywords (e.g., "inventory status", "latest")

[1718] Output: Relevant documents and materials (e.g., documents containing the latest inventory information)

[1719] Specific operation: The server uses a database management system such as SQLite to search for documents and materials that match the keywords. For example, it executes an SQL query to retrieve records that match the keywords.

[1720] Step 4:

[1721] The server sends the search results to the smartphone and displays them to the user.

[1722] Input: Related documents and materials (e.g., documents containing the latest inventory information)

[1723] Output: Search results displayed on the smartphone

[1724] Specific operation: The server converts the search results into a data format such as JSON and sends them to the smartphone. The smartphone application analyzes the received data and displays it in a user-friendly format.

[1725] Step 5:

[1726] The user checks the search results displayed on the smartphone and obtains the necessary information.

[1727] Input: Search results displayed on the smartphone

[1728] Output: Necessary information obtained by the user (e.g., the latest inventory information)

[1729] Specific operation: The user checks the search results displayed on the smartphone screen and obtains the necessary information. For example, check information such as inventory status and delivery schedule.

[1730] (Example 2)

[1731] Next, Example 2 of Embodiment 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset-type terminal 314 is referred to as the "terminal".

[1732] Employees are required to quickly and efficiently search for necessary documents and materials and be provided with results sorted in descending order of relevance. However, in conventional systems, there have been problems such as search results not being properly sorted, which takes time to find the required information. In addition, due to the inappropriate display format of search results, there has been an issue that it is difficult for users to quickly find the required information.

[1733] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following respective means.

[1734] In this invention, the server includes means for searching for relevant documents and materials from an employee's question or keyword using a natural language processing engine, means for sorting the search results in descending order of relevance, and means for generating the search results in a list format. As a result, it becomes possible for employees to quickly and efficiently find the required information.

[1735] The "natural language processing engine" is software for analyzing an employee's question or keyword and searching for relevant documents and materials.

[1736] The "means for searching" is a function for finding documents and materials in a database based on an employee's question or keyword.

[1737] The "means for providing" is a function for displaying search results to an employee.

[1738] The "means for sorting" is a function for sorting search results in descending order of relevance.

[1739] The "means for generating" is a function for formatting search results in a list format.

[1740] The "means for displaying" is a function for displaying the generated search results on an employee's terminal.

[1741] A "machine learning model" is an algorithm that analyzes questions and keywords based on data.

[1742] The "means for categorization" is a function that classifies search results based on specific criteria and automatically stores them in folders or directories corresponding to each category.

[1743] Modes for Carrying Out the Invention

[1744] This invention is a system that enables employees to quickly and efficiently search for necessary documents and materials and provides search results sorted in descending order of relevance. The following describes specific embodiments of this system.

[1745] Configuration of the System

[1746] This system consists of three main elements: a server, a terminal, and a user. The server uses a natural language processing engine and a machine learning model to analyze search queries and search for relevant documents and materials. The terminal is a device for the user to input search queries and display search results. The user is an employee who uses the system to search for necessary information.

[1747] Hardware and Software to be Used

[1748] Server: A computer system equipped with a high-performance processor and a large-capacity memory is used. On the server, Python-based libraries (e.g., NLTK, spaCy) are installed as a natural language processing engine. Also, Elasticsearch is used as the search algorithm.

[1749] Terminal: Devices such as personal computers and smartphones used by the user. A web browser or a dedicated application is installed on the terminal.

[1750] Software: Use machine learning models (e.g., BERT, GPT) to analyze search queries. This enables accurate analysis of user questions and keywords.

[1751] System Operation

[1752] 1. The user enters a search query.

[1753] The user enters the necessary information in the search bar of the terminal. For example, enter "project report".

[1754] 2. The terminal sends the search query to the server.

[1755] The terminal sends the search query entered by the user to the server. The query is sent using an HTTP request.

[1756] 3. The server receives the search query and queries Elasticsearch.

[1757] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch.

[1758] 4. The server receives search results from Elasticsearch.

[1759] The server receives the search results returned by Elasticsearch. The search results contain information on relevant documents and materials.

[1760] 5. The server sorts the search results in descending order of relevance.

[1761] The server sorts the received search results in descending order of relevance. The scoring function of Elasticsearch is used to evaluate relevance.

[1762] 6. The server generates the search results in a list format.

[1763] The server formats the sorted search results into a list in HTML format.

[1764] 7. Transmit the search results generated by the server to the terminal

[1765] The server transmits the generated search results in HTML format to the terminal. The results are transmitted using an HTTP response.

[1766] 8. The terminal displays the search results to the user

[1767] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format.

[1768] 9. The user selects the required documents and materials

[1769] The user selects the required documents and materials from the displayed search results. Click on the selected document to display the details.

[1770] Specific example

[1771] As a specific example, consider the case where a user searches for a "project report". The user enters "project report" from the terminal and clicks the search button. The server receives this query and searches the documents in the database using Elasticsearch. The search results are sorted in descending order of relevance and displayed to the user in a list format. The user can select the required documents from this list.

[1772] Examples of prompt sentences

[1773] Examples of prompt sentences may be as follows.

[1774] Prompt sentence: "Please search for a project report"

[1775] When this prompt text is input into the AI model for generating a project report, the AI model provides search results for quickly finding the project reports required by employees.

[1776] The flow of the specific process in Example 2 will be described with reference to FIG. 13.

[1777] Step 1:

[1778] The user inputs a search query.

[1779] The user inputs the required information into the search bar of the terminal. For example, the user inputs "project report". The input query is temporarily stored in the memory of the terminal.

[1780] Step 2:

[1781] The terminal sends the search query to the server.

[1782] The terminal sends the search query input by the user to the server as an HTTP request. This request contains the query input by the user. After sending the request, the terminal waits for a response from the server.

[1783] Step 3:

[1784] The server receives the search query and queries Elasticsearch.

[1785] The server analyzes the search query received from the terminal and sends a search request to Elasticsearch. Specifically, the server passes the search query to the Elasticsearch API and instructs it to search for relevant documents and materials. The input is the search query, and the output is the search results from Elasticsearch.

[1786] Step 4:

[1787] The server receives the search results from Elasticsearch.

[1788] The server receives the search results returned from Elasticsearch. The search results contain information about relevant documents and materials. The server saves this result in memory to prepare for the next processing. The input is the search results from Elasticsearch, and the output is the search results saved in the server's memory.

[1789] Step 5:

[1790] The server sorts the search results in descending order of relevance.

[1791] The server sorts the received search results in descending order of relevance using Elasticsearch's scoring function. Specifically, it calculates the relevance score for each document and sorts them in descending order. The input is the search results saved in the server's memory, and the output is the sorted search results.

[1792] Step 6:

[1793] The server generates the search results in a list format.

[1794] The server formats the sorted search results into a list in HTML format. Specifically, it encloses the title and summary of each document with HTML tags to make it user-friendly. The input is the sorted search results, and the output is the search results in HTML format.

[1795] Step 7:

[1796] The server sends the generated search results to the terminal.

[1797] The server sends the generated search results in HTML format to the terminal as an HTTP response. The input is the search results in HTML format, and the output is the search results sent to the terminal.

[1798] Step 8:

[1799] The terminal displays the search results to the user.

[1800] The terminal displays the received HTML in a web browser and shows the search results to the user in a list format. Specifically, the web browser parses the HTML and displays it on the screen. The input is the HTML-formatted search results received from the server, and the output is the search results displayed to the user.

[1801] Step 9:

[1802] The user selects the necessary documents and materials.

[1803] The user selects the necessary documents and materials from the displayed search results. Clicking on the selected document displays the details. The input is the user's click operation, and the output is the detailed display of the selected document.

[1804] (Application Example 2)

[1805] Next, Application Example 2 of Form Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset-type terminal 314 is referred to as the "terminal".

[1806] In a logistics center, it is required that employees quickly search for necessary inventory information and delivery information and efficiently perform their operations. However, in the conventional system, there is a problem that it takes time to search for information, and a large amount of irrelevant information is displayed, resulting in a decrease in work efficiency. Furthermore, since information search using smartphones has not been fully utilized, there is also a problem that it is difficult to respond quickly on-site.

[1807] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[1808] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing the documents and materials based on the search results, means for displaying inventory information and delivery information in descending order of relevance, and means for enabling employees to quickly search for necessary information using a smartphone. Thereby, employees can quickly search for necessary information and improve work efficiency.

[1809] The "natural language processing engine" is software for analyzing employees' questions and keywords and searching for relevant documents and materials.

[1810] The "means for providing search results" is a function for displaying the searched documents and materials to employees.

[1811] The "means for organizing documents and materials" is a function for categorizing documents and materials based on the search results and automatically storing them in folders or directories corresponding to each category.

[1812] The "means for displaying inventory information and delivery information in descending order of relevance" is a function for sorting and displaying inventory data and delivery data in descending order of relevance based on a search query.

[1813] The "means for enabling employees to quickly search for necessary information using a smartphone" is a function for enabling employees to quickly search for necessary information using a smartphone.

[1814] The system for implementing this invention enables employees to quickly search for necessary inventory information and delivery information in a logistics center. The following describes specific embodiments of this system.

[1815] Configuration of the System

[1816] This system is composed of the following main components:

[1817] 1. Server: Equipped with a natural language processing engine and a machine learning model, it analyzes employees' questions and keywords.

[1818] 2. Smartphone: A terminal used by employees to input search queries and display search results.

[1819] 3. Database: Stores inventory information and delivery information.

[1820] Program Processing

[1821] The server operates as follows:

[1822] 1. Natural Language Processing Engine: Analyzes questions and keywords input by employees from smartphones and searches for relevant documents and materials.

[1823] 2. Machine Learning Model: Generates search results in descending order of relevance based on the analysis of questions and keywords.

[1824] 3. Providing Search Results: Sends the search results to the smartphone and provides them to the employees.

[1825] 4. Data Sorting: Categorizes documents and materials based on the search results and automatically stores them in the corresponding folders and directories for each category.

[1826] 5. Displaying Inventory Information and Delivery Information: Sorts and displays inventory data and delivery data in descending order of relevance.

[1827] Hardware and Software Used

[1828] Hardware: Smartphones, servers

[1829] Software: Python, Pandas, Scikit-learn, natural language processing engine (e.g., SpaCy)

[1830] Specific Example

[1831] When an employee searches for "products out of stock", the system operates as follows:

[1832] 1. The employee enters "products out of stock" into the smartphone.

[1833] 2. The natural language processing engine of the server analyzes this query and searches for relevant inventory information.

[1834] 3. The machine learning model generates search results in descending order of relevance.

[1835] 4. The search results are displayed on the smartphone, and the employee can view the list of products out of stock.

[1836] Example of Prompt Sentence

[1837] Search Query: "products out of stock"

[1838] In this way, employees in the logistics center can quickly search for the necessary information and improve work efficiency.

[1839] The flow of specific processing in Application Example 2 will be described with reference to FIG. 14.

[1840] Step 1:

[1841] The user enters a search query into the smartphone.

[1842] Input: The user enters "products out of stock" into the smartphone.

[1843] Output: The search query is sent to the server.

[1844] Specific Operation: The user enters "products out of stock" into the search bar of the smartphone and presses the search button.

[1845] Step 2:

[1846] The server analyzes the search query using a natural language processing engine.

[1847] Input: Search query "out-of-stock products" sent from a smartphone.

[1848] Output: Analyzed query data.

[1849] Specific operation: The server's natural language processing engine (e.g., SpaCy) tokenizes the search query and extracts important keywords.

[1850] Step 3:

[1851] The server searches for relevant documents and materials using a machine learning model.

[1852] Input: Analyzed query data.

[1853] Output: List of relevant documents and materials.

[1854] Specific operation: The server's machine learning model (e.g., Scikit-learn) uses the analyzed query data to search for inventory information and delivery information in the database and sorts them in descending order of relevance.

[1855] Step 4:

[1856] The server sends the search results to the smartphone.

[1857] Input: List of relevant documents and materials.

[1858] Output: Search results displayed on the smartphone.

[1859] Specific operation: The server sends the list of relevant documents and materials to the smartphone and displays them to the user.

[1860] Step 5:

[1861] The user checks the search results on the smartphone.

[1862] Input: Search results displayed on the smartphone.

[1863] Output: The user checks the required information.

[1864] Specific operation: The user scrolls through the search results displayed on the smartphone screen and checks the required inventory information and delivery information.

[1865] Step 6:

[1866] The server organizes documents and materials based on the search results.

[1867] Input: List of relevant documents and materials.

[1868] Output: Categorized documents and materials.

[1869] Specific operation: The server categorizes documents and materials based on the search results and automatically stores them in the corresponding folders or directories for each category.

[1870] Step 7:

[1871] The server displays the inventory information and delivery information in descending order of relevance.

[1872] Input: Relevant inventory information and delivery information.

[1873] Output: Inventory information and delivery information sorted in descending order of relevance.

[1874] Specific operation: The server sorts the inventory data and delivery data in descending order of relevance and displays it to the user.

[1875] (Example 3)

[1876] Next, Example 3 of Embodiment 3 will be described. In the following description, the data processing device 12 is referred to as a "server", and the headset type terminal 314 is referred to as a "terminal".

[1877] There is a problem that it is difficult for employees to efficiently search for and organize the necessary information. Especially when there are a large number of documents and materials, manual search and organization require time and effort, and the work efficiency decreases. Also, when the search results are not appropriately categorized, it is difficult to quickly find the necessary information.

[1878] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 3 is realized by the following means. In this invention, the server includes means for searching for relevant information from the questions and keywords of employees using a natural language processing engine, means for providing the search results to employees, means for analyzing and categorizing the information based on the search results, and means for automatically storing the categorized information in folders or directories corresponding to each category. Thereby, employees can efficiently search for and organize the necessary information.

[1879] The "natural language processing engine" is software for analyzing natural language and understanding its meaning.

[1880] An "employee" is an individual who belongs to a company or organization and conducts business.

[1881] "Questions and keywords" are input data used by employees when searching for information.

[1882] "Relevant information" is documents and materials searched based on the questions and keywords of employees.

[1883] "Means for searching" is a method or technology for obtaining relevant information based on the questions and keywords of employees.

[1884] "The means of providing" refers to the methods and technologies for displaying search results to employees.

[1885] "The means of analyzing and categorizing" refers to the methods and technologies for analyzing search results and classifying them into specific categories.

[1886] "Folders and directories" refer to virtual storage locations for organizing information within a computer.

[1887] "The means of automatically storing" refers to the methods and technologies for automatically moving the analyzed information to the corresponding folders and directories.

[1888] "Machine learning model" refers to an algorithm that learns based on data and performs predictions and classifications.

[1889] This invention is a system for employees to efficiently search for and organize the necessary information. The following describes the specific embodiments of this system.

[1890] Configuration of the system

[1891] This system is composed of three main elements: a server, a terminal, and a user.

[1892] 1. Server:

[1893] The server uses a natural language processing engine to search for relevant information from employees' questions and keywords. Specifically, it uses search engines such as Apache Solr and Elasticsearch. Also, it uses natural language processing tools such as Google Cloud Natural Language API and IBM Watson Natural Language Understanding to analyze the search results and perform categorization. Furthermore, it uses Python's os module and shutil module to automatically store the categorized information in the corresponding folders and directories.

[1894] 2. Terminal:

[1895] The terminal is a device for the user to input a search query and view search results. The terminal consists of computer devices such as a personal computer, tablet, smartphone, etc.

[1896] 3. User:

[1897] The user is an employee who uses the system to search for and organize information. The user inputs a search query using the terminal and views the organized information.

[1898] System Operation

[1899] 1. Input of search query by user:

[1900] The user inputs a question or keyword into the search bar of the terminal. For example, input "2023 annual business report".

[1901] 2. Obtaining search results by server:

[1902] The server receives the search query input by the user and uses Apache Solr or Elasticsearch to obtain relevant information from the database.

[1903] 3. Analysis and categorization of search results by server:

[1904] The server analyzes the obtained search results using Google Cloud Natural Language API or IBM Watson Natural Language Understanding and classifies them into appropriate categories. For example, classify them into categories such as "business report", "financial report", "market analysis", etc.

[1905] 4. Storage of each category into folders by server:

[1906] The server automatically stores the categorized information into folders and directories corresponding to each category using Python's os module and shutil module. For example, the information classified into the "Business Report" category is stored in the "Business Report" folder.

[1907] 5. Confirmation of the organized information by the user:

[1908] The user checks the organized folders using the terminal. For example, the user can open the "Business Report" folder and view the necessary information.

[1909] Examples of specific cases and prompt sentences

[1910] Specific case:

[1911] The specific actions when the user searches for the "Business Report for 2023" are as follows.

[1912] 1. User: Enter "Business Report for 2023" in the search bar of the terminal and click the search button.

[1913] 2. Server: Receive the search query and use Apache Solr to retrieve relevant information from the database.

[1914] 3. Server: Analyze the retrieved information using the Google Cloud Natural Language API and classify it into the "Business Report" category.

[1915] 4. Server: Move the classified information to the "Business Report" folder using Python's os module.

[1916] 5. User: Open the "Business Report" folder on the terminal and check the necessary information.

[1917] Example of a prompt sentence:

[1918] "Search for the 2023 business report and automatically store the relevant information in the business report folder."

[1919] With this system, users can efficiently search for and organize the necessary information. The flow of the specific process in Example 3 will be described with reference to Figure 15.

[1920] Step 1: Input of search query by the user

[1921] The user enters a question or keyword in the search bar of the terminal. For example, enter "2023 business report". The input data is the text related to the information the user wants to search for. The output is sent to the server as a search query.

[1922] Step 2: Obtaining search results by the server

[1923] The server receives the search query entered by the user and uses a search engine (e.g., Apache Solr or Elasticsearch) to obtain relevant information from the database. The input data is the user's search query. The server sends the search query to the search engine and obtains relevant documents and materials. The output is a list of documents and materials as search results.

[1924] Step 3: Analysis and categorization of search results by the server

[1925] The server analyzes the obtained search results using a natural language processing engine (e.g., Google Cloud Natural Language API or IBM Watson Natural Language Understanding). The input data is the documents and materials obtained as search results. The server analyzes these documents and classifies them into appropriate categories. For example, classify them into categories such as "business report", "financial report", "market analysis", etc. The output is a list of categorized documents and materials.

[1926] Step 4: Storage in Folders for Each Category by the Server

[1927] The server automatically stores the categorized documents and materials in the folders or directories corresponding to each category. The input data is the categorized documents and materials. The server uses the os module and shutil module in Python to obtain the document paths and move them to the corresponding folders. For example, the documents classified into the "Sales Report" category are stored in the "Sales Report" folder. The output is the documents and materials stored in the folders.

[1928] Step 5: Confirmation of the Organized Information by the User

[1929] The user checks the organized folders using the terminal. The input data is the documents and materials stored in the folders. The user can open, for example, the "Sales Report" folder and view the necessary documents. The output is the documents and materials viewed by the user.

[1930] With this system, the user can efficiently search for and organize the necessary information.

[1931] (Application Example 3)

[1932] Next, Application Example 3 of Embodiment Example 3 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset type terminal 314 is referred to as the "terminal".

[1933] In the conventional document management system, the search and organization of documents and materials are often performed manually, which has the problem of low efficiency. Also, at sites such as logistics centers, there are many paper-based documents, and means for digitizing and efficiently managing them have been demanded. Furthermore, the categorization of documents and the upload to cloud storage are often performed manually, which has the problem of taking time and labor.

[1934] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 3 is realized by the following means.

[1935] In this invention, the server includes means for searching for relevant documents and materials from employees' questions and keywords using a natural language processing engine, means for providing the search results to employees, means for organizing documents and materials based on the search results, means for extracting text from images using optical character recognition technology, means for categorizing documents based on the extracted text and automatically storing them in corresponding folders, and means for uploading documents to cloud storage. Thereby, efficient search, organization, digitization, and automatic upload to cloud storage of documents and materials become possible.

[1936] The "natural language processing engine" is software for analyzing natural language and searching for relevant information from questions and keywords.

[1937] "Employees" refer to people who belong to a company or organization and perform their duties.

[1938] The "means for searching" refers to methods and technologies for finding specific information.

[1939] The "means for providing" refers to methods and technologies for displaying search results to users or making them accessible.

[1940] The "means for organizing" refers to methods and technologies for classifying and managing documents and materials based on specific criteria.

[1941] "Optical character recognition technology" is a technology for extracting character information from images.

[1942] "Image" refers to a digital file containing visual information.

[1943] "Text" refers to digital data containing character information.

[1944] "Categorization" refers to classifying information based on specific criteria.

[1945] "Folder" refers to a virtual container for organizing digital data.

[1946] "Cloud storage" refers to a remote server for storing data through the Internet.

[1947] "Upload" refers to transferring data from a local device to a remote server.

[1948] The system for implementing this invention is configured as follows. First, the server uses a natural language processing engine to search for relevant documents and materials from employees' questions and keywords. The natural language processing engine analyzes the questions and keywords based on a machine learning model. The analyzed results are provided to the employees.

[1949] Next, the server organizes the documents and materials based on the search results. As a means of organization, optical character recognition technology (OCR) is used to extract text from images. The extracted text is categorized based on specific keywords and automatically stored in the corresponding folders. Furthermore, the documents are uploaded to cloud storage.

[1950] To realize this system, the following hardware and software are used. As hardware, a smartphone with a camera is required. As software, Python, an OCR library, PIL (Python Imaging Library), and Google Cloud Storage are used.

[1951] As a specific example, consider a document organization app used in a logistics center. When an employee takes a photo of a shipping instruction document with a smartphone camera, the image is converted into text using OCR technology. The converted text is classified into the category of "shipping instruction document" and automatically stored in the corresponding folder. Subsequently, the document is uploaded to cloud storage.

[1952] Examples of prompt texts may include the following.

[1953] "Please develop a document organization app for use in a logistics center. It is an application that analyzes document images taken with a smartphone camera using OCR technology, automatically classifies them into categories such as shipping instruction documents, receiving documents, and inventory lists, and has the function of uploading them to cloud storage."

[1954] In this way, efficient search, organization, digitization of documents and materials, and automatic upload to cloud storage become possible.

[1955] The flow of the specific process in Application Example 3 will be described with reference to FIG. 16.

[1956] Step 1:

[1957] The user takes a photo of a document with a smartphone camera.

[1958] Input: Paper-based document

[1959] Output: Digital image file

[1960] Specific operation: The user launches the smartphone camera app and takes a photo of the document. The captured image is saved in the smartphone.

[1961] Step 2:

[1962] The terminal extracts text from the image using optical character recognition technology (OCR).

[1963] Input: Digital image file

[1964] Output: Extracted text data

[1965] Specific operation: The terminal inputs the saved image file into OCR software and extracts the character information in the image as text data.

[1966] Step 3:

[1967] The terminal analyzes the extracted text and categorizes it.

[1968] Input: Extracted text data

[1969] Output: Category information

[1970] Specific operation: The terminal analyzes the extracted text data and determines the category based on specific keywords (e.g., "shipping instruction", "receipt", etc.).

[1971] Step 4:

[1972] The terminal automatically stores the document in the corresponding folder based on the category.

[1973] Input: Category information, digital image file

[1974] Output: Document stored in the folder

[1975] Specific operation: The terminal moves or copies the digital image file to the corresponding folder based on the determined category.

[1976] Step 5:

[1977] The terminal uploads the document to the cloud storage.

[1978] Input: Document stored in the folder

[1979] Output: Documents stored in cloud storage

[1980] Specific operation: The terminal uploads the documents in the folder to a cloud storage service (Google Cloud Storage). When the upload is complete, the documents are stored on the cloud.

[1981] Step 6:

[1982] The server uses a natural language processing engine to search for relevant documents and materials from the employee's questions and keywords.

[1983] Input: Employee's questions and keywords

[1984] Output: Search results of relevant documents and materials

[1985] Specific operation: The server inputs the questions and keywords input by the employee into a natural language processing engine to search for relevant documents and materials. The search results are provided to the employee.

[1986] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform specific processing using the user's emotion.

[1987] "Form Example 1"

[1988] In one embodiment of the present invention, a natural language processing engine and an emotion engine are combined.

[1989] When an employee enters a question or keyword into the system, the natural language processing engine analyzes this keyword and searches the database for relevant documents and materials. At the same time, the sentiment engine analyzes the emotional context of the employee's question or keyword. For example, when an employee enters the phrase "Can't find the latest sales report", the sentiment engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[1990] "Morphological Example 2"

[1991] In another embodiment of the present invention, the sentiment engine sorts documents and materials based on the emotional context of the employee. Specifically, the sentiment engine analyzes the emotional context of the employee and automatically stores the documents and materials in the folders or directories corresponding to each emotional context. For example, when an employee enters the phrase "Can't find the sales report", the sentiment engine recognizes the employee's frustration from this phrase, and as a result, the system automatically stores the relevant documents and materials in the "Frustration" folder.

[1992] "Morphological Example 3"

[1993] In one embodiment of the present invention, the natural language processing engine and the sentiment engine are combined.

[1994] When an employee enters a question or keyword into the system, the natural language processing engine analyzes this keyword and searches the database for relevant documents and materials. At the same time, the sentiment engine analyzes the emotional context of the employee's question or keyword. For example, when an employee enters the phrase "Can't find the latest sales report", the sentiment engine recognizes the employee's frustration from this phrase. This information is used for the system to provide more relevant documents and materials.

[1995] The processing flow of each exemplary embodiment will be described below.

[1996] "Exemplary Embodiment 1"

[1997] Step 1: An employee inputs questions or keywords to the system.

[1998] Step 2: The natural language processing engine analyzes this keyword and searches for relevant documents and materials from the database.

[1999] Step 3: The sentiment engine analyzes the emotional context of the employee's questions or keywords.

[2000] Step 4: The natural language processing engine utilizes the analysis result of the sentiment engine to provide more relevant documents and materials.

[2001] "Exemplary Embodiment 2"

[2002] Step 1: An employee inputs questions or keywords to the system.

[2003] Step 2: The sentiment engine analyzes the emotional context of the employee's questions or keywords.

[2004] Step 3: Based on the analysis result of the sentiment engine, documents and materials are automatically stored in folders or directories corresponding to each emotional context.

[2005] (Example 1)

[2006] Next, Example 1 of Exemplary Embodiment 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the headset-type terminal 314 is referred to as the "terminal".

[2007] In a conventional system, when searching for relevant documents and materials in response to questions or keywords input by employees, it is impossible to consider the emotional context, making it difficult to fully meet the needs of employees. Also, in the organization and provision of search results, appropriate information provision that reflects the emotions of employees is not carried out, thus hindering efficient business operations.

[2008] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in the first embodiment is realized by the following respective means.

[2009] In this invention, the server includes means for searching for relevant documents and materials from the questions and keywords of employees using a natural language processing engine, means for providing the search results to the employees, means for organizing the documents and materials based on the search results, means for analyzing the emotional context of the questions and keywords of employees using an emotion engine, and means for providing highly relevant documents and materials based on the emotional context. Thereby, appropriate information provision considering the emotions of employees becomes possible, enabling the improvement of business efficiency and employee satisfaction.

[2010] The "natural language processing engine" is software for analyzing questions and keywords input by employees, understanding their meanings, and searching for relevant documents and materials.

[2011] The "machine learning model" is an algorithm that learns based on data and finds patterns and rules, and is used to improve the analysis ability of the natural language processing engine.

[2012] The "emotion engine" is software for analyzing the emotional context included in the questions and keywords of employees and recognizing their emotions.

[2013] "Documents and materials" is a general term for texts, reports, presentations, datasheets, etc. that contain information necessary for employees to perform their work.

[2014] A "database" is an information system for efficiently storing, retrieving, and managing documents and materials.

[2015] The "search results" are a list of documents and materials retrieved from the database based on questions and keywords analyzed by a natural language processing engine and an emotion engine.

[2016] "Categorization" is a process of classifying documents and materials according to specific themes or attributes based on the search results.

[2017] "Folders and directories" are virtual storage locations on a computer for organizing and storing documents and materials.

[2018] An "employee" is a user who inputs questions and keywords to perform tasks using the system.

[2019] Mode for Implementing the Invention

[2020] This invention is a system that analyzes questions and keywords input by employees and provides relevant documents and materials. The system combines a natural language processing engine and an emotion engine to provide information considering the emotional context of employees.

[2021] Hardware and Software to be Used

[2022] Server

[2023] The server performs the main processing of the system using the following software:

[2024] Natural language processing engine (e.g., BERT, GPT-3)

[2025] Emotion engine (e.g., emotion analysis API)

[2026] Database management system (e.g., MySQL, PostgreSQL)

[2027] Terminal

[2028] The terminal provides an interface for employees to enter questions and keywords. The terminal uses the following software:

[2029] Web browser or dedicated application

[2030] ...

Claims

1. means for analyzing keywords or meanings from the user's input information using a natural language processing engine; means for analyzing the emotional context of the user using an emotion engine including an emotion identification model that uses a neural network trained with emotion values indicating a plurality of emotions mapped to an emotion map; means for searching for relevant document data by combining the analysis result by the natural language processing engine and the analysis result by the emotion engine; means for sorting the search results of the document data in descending order of relevance according to the emotional context of the user; means for providing the sorted search results to the user; means for categorizing the document data based on the emotional context of the user and storing the categorized document data corresponding to the emotional context; A system comprising.

2. The system according to claim 1, further comprising means for providing a detailed explanation or video according to the emotional context of the user. The system according to claim 1.

Images