Device and method

Abstract

The objective of the present disclosure is to provide a device and a method enabling presentation of an analysis method based on person-dependent knowledge.　In a prompt generation device 100 according to the present disclosure, a terminal access unit 101 functions as a reception unit and receives request information for data processing from a user. In the present disclosure, the data processing is processing for SQL search but obviously may also include other processing. Then, an acquisition unit 102 acquires related information related to the data processing on the basis of the request information. A prompt generation unit 103 generates a prompt for instructing a generative AI model to generate a proposal content for the data processing on the basis of the related information.

Description

Apparatus and method

[0001] The present invention relates to an apparatus and method for utilizing a language model.

[0002] Patent Document 1 describes that, in order to make it easier for users who do not have knowledge of databases to analyze data, the specification of analysis items indicating the content of the analysis is accepted, and multiple analysis methods are displayed according to the attributes of the target table, which is the table corresponding to the analysis items.

[0003] Japanese Patent Application Laid-Open No. 2020-170371

[0004] However, data processing, such as data analysis methods, requires expert knowledge that is highly dependent on individuals, and the technology described in Patent Literature 1 makes it difficult to present an analysis method that corresponds to such individualized knowledge. For example, it is unable to address knowledge about in-house data, such as what data is available within a company and which data should be combined and used.

[0005] Therefore, an object of the present disclosure is to provide an apparatus and method that can present a data processing method based on personalized knowledge.

[0006] The device disclosed herein includes a reception unit that receives request information for data processing from a user, an acquisition unit that acquires related information related to the data processing based on the request information, and a generation unit that generates a prompt to instruct a generation AI to generate suggested content for the data processing based on the related information.

[0007] According to the present disclosure, it is possible to present a data processing method based on personalized knowledge.

[0008] FIG. 1 is a diagram showing the system configuration of a search system including a prompt generation device 100 according to the present disclosure. FIG. 2 is a block diagram showing the functional configuration of the prompt generation device 100. FIG. 3 is a flowchart showing the operation of the prompt generation device 100 according to the present disclosure. FIG. 4 is a diagram showing a subroutine illustrating detailed processing of step S103. FIG. 5 is a diagram showing a specific example of a prompt. FIG. 6 is a diagram showing a specific example of an answer generated by the LLM 300. FIG. 7 is a diagram showing various databases. FIG. 8 is a diagram showing an example of the hardware configuration of the prompt generation device 100 according to an embodiment of the present disclosure.

[0009] The present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0010] 1 is a diagram showing the system configuration of a search system including a prompt generation device 100 according to the present disclosure. As shown in the diagram, the prompt generation device 100 can be connected to and communicate with an in-house DB 200, an LLM 300, and a user terminal 400 via a network.

[0011] When the prompt generation device 100 receives request information for data processing from a user terminal, it accesses the in-house DB 200 to retrieve data related to the request information for data processing. In this disclosure, the in-house DB 200 is a database that stores information such as SQL queries for data processing used within a company. In this disclosure, the in-house DB 200 is a database that is external to the prompt generation device 100 and is treated as an external DB.

[0012] Note that all or some of the LLM 300 and the in-house DB 200 may be located in the prompt generation device 100 or the user terminal 400. Furthermore, the user terminal 400 may have the functions of the prompt generation device 100 and function as the prompt generation device 100. There are types of generation AI models, such as "tsuzumi," in which the generation AI model is located inside the user terminal 400. In this type, the RAG app is also provided on the user terminal 400. However, the information accessed by the RAG app (knowledge DB) may be located inside the user terminal 400 or on the network. There are also types, such as ChatGPT, in which the generation AI model is located on the network. In this type, the RAG app is provided on the user terminal 400. However, the information accessed by the RAG (knowledge DB) is located on the network.

[0013] For example, a user terminal 400 equipped with Tsuzumi has a prompt generation function (corresponding to the prompt generation device 100), a generation AI (corresponding to the LLM 300), an RAG application (RAG function), and a knowledge DB (corresponding to an external DB). When a user operates the user terminal 400, the prompt generation function executes the RAG function, accesses the knowledge DB, retrieves the necessary information, generates a prompt, outputs it to the built-in generation AI, and obtains the answer. The user terminal 400 does not need to have a built-in knowledge DB, and may instead access a knowledge DB on the network to retrieve the necessary information.

[0014] A generative AI model, such as the LLM 300, is a model that generates content in response to a prompt containing input information, based on the instructions, context, question, and output format indicated by the prompt, and returns the content as response information. The prompt can also contain input information, in which case the generative AI model generates response information targeted at the input information. The generative AI model may be, for example, an interactive AI model that includes a large language model (LLM) and a user interface (UI) for interacting with the user, enabling text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, PaLM2, and the like. In this embodiment, the prompt generation device 100 is capable of providing content provision functions using interactive AI models, which are multiple types of models. These interactive AI models may be stored within the prompt generation device 100, or may be stored in another device connected to the prompt generation device 100 via a network so that information can be exchanged with the user via the prompt generation device 100. Although only one prompt generation device 100 is shown in the figure, multiple prompt generation devices 100 may be included.

[0015] In this disclosure, a prompt is information indicating an instruction or question entered by a user in an interactive system such as an interaction with a generative AI model or a command line interface (CLI).

[0016] The prompt generation device 100 then generates a prompt that includes instructions for the LLM 300 to generate an answer that meets the user's request, based on the extracted data. The prompt generation device 100 then transmits the prompt to the LLM 300, thereby obtaining an answer corresponding to the request information from the LLM 300. The prompt generation device 100 then transmits the obtained answer to the user terminal 400, allowing the user of the user terminal 400 to obtain an answer to the request information.

[0017] 2 is a block diagram showing the functional configuration of the prompt generation device 100. The prompt generation device 100 includes a terminal access unit 101, an acquisition unit 102, a prompt generation unit 103, and an LLM access unit 104.

[0018] The terminal access unit 101 is a unit that receives request information from the user terminal 400. The terminal access unit 101 also receives an LLM response from the LLM access unit 104 and transmits it to the user terminal 400.

[0019] The acquisition unit 102 is a part that acquires a data processing method (related information) related to the request information based on the request information. More specifically, the acquisition unit 102 first searches the query DB (query information) stored in the in-house DB 200 based on the request information. The acquisition unit 102 determines the top n search queries based on the similarity between each search query (data record unit) included in the query DB (query information) and the request information. In the present disclosure, the search query includes information such as the date the query was submitted, the query creator, the SQL statement, the table name (including the column name and the view name), whether the query is correct, etc., but is not limited to these.

[0020] The prompt generator 103 acquires table names, column names, view names, etc. from the n acquired search queries and generates a prompt based on them. This prompt includes information that allows the LLM 300 to propose an analysis method based on the request information requested by the user, and the LLM 300 generates and returns a response to that information.

[0021] The LLM access unit 104 is a unit that sends the generated prompt to the LLM 300 and obtains the response.

[0022] The operation of the prompt generation device 100 configured as above will now be described. Fig. 3 is a flowchart showing the operation of the prompt generation device 100 of the present disclosure. The terminal access unit 101 receives request information from the user terminal 400 (S101). The acquisition unit 102 then searches the in-house DB 200 for a search query based on the request information (S102). The acquisition unit 102 acquires at least one search query that conforms to the request information (S103).

[0023] The prompt generation unit 103 obtains the table name, column name, view name, SQL statement, searcher, etc. from each obtained search query (S104), and generates a prompt based on the table name, column name, view name, etc. (S105).

[0024] The LLM access unit 104 transmits the generated prompt to the LLM 300 and obtains the response generated by the LLM 300 (S106).

[0025] 4 is a diagram showing a subroutine illustrating detailed processing of step S103. The acquisition unit 102 vectorizes the request information (S103a). The acquisition unit 102 compares the similarity between the request information and the query information (S103b). This similarity is calculated based on the vector of the request information and the vector of the search query. The acquisition unit 102 acquires the top n search queries whose similarity is equal to or greater than a predetermined value (S103c).

[0026] This allows the user to use the in-house DB 200 to obtain a personalized search query and use it to have the LLM 300 propose an appropriate analysis method.

[0027] Next, a specific example of a prompt generated from request information in this disclosure and an answer generated in the LLM 300 based on the prompt will be described. Figure 5 is a diagram showing a specific example of a prompt. As shown in the figure, the prompt includes a role, a task, and a condition.

[0028] 5, the role column states, "You are a DB specialist." This prompt requests the LLM 300 to identify as a DB specialist.

[0029] The task field also states, "Please propose the tables and analysis methods to be used, as well as SQL statements based on them. Please also include the intention behind using the tables and any other tables considered. Please use the examples as a reference when making your proposal." The task field contains predetermined, standardized information. The text information written in this task field is an example, and naturally, other formats of information may be used. In the present disclosure, as written in this task field, the user is required to propose to the LLM 300 the table names to be used in data processing, the analysis methods, and SQL statements based on those, as well as the reasons for the proposals.

[0030] The condition field contains the user's request information, such as "I want to analyze the subscriber transition of members." The condition field also contains the table names, column names (xx_table, cccccc, yy_table, yyyy, zz_table, zzzz, aa_table, aaaa), view names (xx_view), and SQL statements (SQL statements (1) and (2)) of various tables contained in each search query acquired by the acquisition unit 102 from the query information acquired from the internal DB 200. The information shown in the condition field is treated as examples by the LLM 300, and the LLM 300 can select tables to use based on this information.

[0031] The acquired search query also includes database operation instruction information, such as SQL statements (1) and (2), although the specific description is simplified in the figure.

[0032] The person instructing the search query is also included as a person in charge. In this case, it is written as Search Taro, and the search query searched by Search Taro is requested. As will be described later, this means that a person who searches frequently (above a certain frequency) is acquired as the search instructor and reflected in the prompt, but this is not necessary.

[0033] The output format is specified as a table name, the reason for adopting it, and an SQL statement. This output format is predetermined standard information.

[0034] FIG. 6 shows a specific example of a response generated by the LLM 300. As shown in the figure, search content 1 shows the required table names (e.g., xx_table, yy_table, zz_table) and the reasons for their rejection. Other considerations show the names of tables considered (e.g., aa_table) and the reasons for their rejection. Proposal content 2 shows an SQL statement, indicating database operation instruction information for the request. The operation instructor corresponds to the person in charge included in the prompt.

[0035] When the user inputs request information into the user terminal 400 (sends it to the prompt generation device 100), the prompt shown in FIG. 5 is generated, and in response to that prompt, the LLM 300 can generate the answer shown in FIG. 6.

[0036] 5 and 6 are merely examples, and other information may be included. For example, the analysis method generated in the LLM 300 may be instructed to include the information shown below.

[0037] For example, the prompt may include information instructing which database to select. For example, the database to be used can be determined based on the content of the request information. For example, if the request is for the transition of membership information, the prompt may instruct the user to select a database that contains the membership information.

[0038] Furthermore, if there is an SQL statement appropriate for a particular analysis (visualization method (pie chart, bar graph)), the content may also include information indicating that statement. If a visualization method is included in a search query that matches the request information from the in-house DB 200 (query DB, etc.), the acquisition unit 102 may extract the visualization method.

[0039] The database operations may include, for example, searching (select), registering (insert), updating (update), and deleting (delete). The acquisition unit 102 can extract this information from the SQL statements included in the query DB.

[0040] The information may include the type of analysis method. For example, the information may include a visualization method (pie chart, bar graph), or information about the output format (for example, what key to use and in what order to display, and whether to highlight data that meets a predetermined condition (font manipulation such as enlarging the character size, coloring characters, underlining, or bolding)). The acquiring unit 102 can acquire this information if it is included in the query DB.

[0041] The query DB may include information on what table name and what view name to use. Multiple view names may be specified. For example, in the case of a subquery, the view name to be used in the subquery and the view name to be used in the main query may be proposed. The acquisition unit 102 can extract this information from the query DB.

[0042] In addition, it may include information used for DB operations (for example, column names, conditional expressions (if statements, when statements)), etc.

[0043] As described above, the information to be included in the analysis method is information extracted from the in-house DB 200 and is determined from information included in each search query extracted based on vector similarity.

[0044] The type of analysis method may be determined, for example, as follows.

[0045] As an analysis method using the user's usage history, the text information included in the request information may be compared with the in-house DB 200 (information of examples), and the corresponding examples (parts of them) may be output to the prompt, as described above.

[0046] The prompt generation device 100 may also determine an analysis method in the LLM 300 by using SQL statements previously used by the user, which are stored in the in-house DB 200. By accessing the in-house DB 200, the prompt generation device 100 can obtain SQL statements previously used by the user.

[0047] For example, as shown in Example 1 below, past SQL statements and analysis methods (e.g., visualization method, analysis purpose) may be stored in the in-house DB 200, and examples of how each analysis content (e.g., theme (e.g., d-payment), target information (e.g., payment information), and analysis purpose) was visualized may be used. Example 1 Theme Information Analysis purpose Visualization method "Point payment, payment information, composition ratio analysis, pie chart" "Point payment, user trend analysis, bar graph" Also, as shown in Examples 2 and 3 below, past SQL statements and themes may be recorded, and examples may be used in which table names, view names, and column names corresponding to themes to be analyzed are associated and saved. Furthermore, frequency of use may be associated and saved, and the top few results may be output to a prompt. Example 2: Theme Table name or view name ("Point payment, xx_table", "Point payment xx_view") Example 3: Theme Table name or view name, column name ("Point payment, xx_table, cccccc") Furthermore, as shown in Example 4 below, conditional expressions used in past SQL statements may be stored as examples for each variation, theme, and analysis purpose. The frequency of use may be associated and saved, and the top few results may be output to the prompt. Example 4: Point payment, trend analysis of user numbers, conditional expression indicating a usage amount of 5,000 yen or more For example, the analysis content and the SQL statement itself may be associated and saved in the internal DB 200, and this may be saved as an example. The frequency of use may be associated and saved, and the top few results may be listed. Example 5: Point payment, trend analysis of user numbers, SQL statement 1 Furthermore, in addition to Examples 1 to 5 above, information on the department or person who submitted the SQL statement may be associated and stored, and this may be output. Only user information whose submission count (search frequency) exceeds a predetermined threshold may be registered. Example 6: Point payment, analysis of trends in number of users, Intellectual Property Department, Search Taro. In addition, the present disclosure may include the following processing.

[0048] Generally, as shown in Figure 7(a), the in-house DB 200 stores a query DB that includes the date the query was submitted, who (email address), the SQL query written by that person, a list of table names that appear in the SQL query, and whether they are correct as a query (whether any errors have occurred).

[0049] As other data, as shown in FIG. 7(b), the in-house DB 200 stores a table DB including "table name, summary of data contained in the table (e.g., line contract information at a certain point in time for each basic contract line of the mobile phone business is made public. Various contract information of line subscribers (including cancellations) can be obtained), update cycle, retention period, and retention start date."

[0050] 7(c), the in-house DB 200 may store a history DB (past history) by combining the query DB and table DB described above, including "date, who (email address only), SQL query written by that person, list of table names appearing in the SQL query, what kind of tables those were, and whether those are correct as queries (whether any errors occurred)." The acquisition unit 102 may acquire various information for generating prompts from this history DB.

[0051] Furthermore, by using the LLM 300 or other LLMs, the search query (SQL query) written by the person can be rewritten in natural language to explain what in-house DB 200 was used and what kind of analysis was performed.

[0052] For example, a search query can be created that uses a conditional expression to extract only users who spend more than 5,000 yen per month from a table of user A's membership information and user B's point payment history information, and then outputs the monthly point payment amounts for those users in 2023.

[0053] Naturally, the data may not be expressed in natural language, but may be obtained in an organized form such as: - Which table was used? - User A's Docomo membership information table - User B's point payment history information - Under what conditions - A condition that extracts only users who spend more than 5,000 yen per month - What was output? - The user's monthly d payment usage amount for 2023 was output.

[0054] By rewriting past search queries, which have different content, items, and description methods, into natural language using LLM, it is possible to compare them from the same perspective.

[0055] In addition, since the query DB contains user attribute information such as email addresses that can identify individuals, the prompt generation device 100 can obtain the number of times a person has written SQL queries (measuring their level of expertise) by referring to the photo DB.

[0056] Furthermore, it is possible to obtain a combination of information related to themes (e.g., what themes or projects the employee has experience in), departments, and DB operation experience, which are linked to the email address, from the employee DB and store the information in the in-house DB 200. Fig. 7D is a diagram showing a specific example of the employee DB.

[0057] Using this, you can also obtain the following information: - Which table was used? - User A's membership information table - User B's point payment history information - Under what conditions - Condition to extract only users who spend more than 5,000 yen per month - What was output? - Outputted the user's monthly point payment amount for 2023 Not only this, you can also obtain information such as: - What department users are working on similar queries - What kind of background users are working on similar queries - What project similar queries are being worked on.

[0058] The information obtained here allows comparison not only of the similarity of the desired analysis theme but also of personal experience. It is also possible to prioritize search queries to be output based on skill and suggest similar queries from users who frequently write search queries.

[0059] In the above disclosure, the graph was determined by the LLM 300 as one of the analysis methods, but this is not the only option. The graph can be treated as a visualization method rather than an analysis method. In this case, the visualization method can be determined based on information in the request information, such as "what you want to see using the data (e.g., annual purchase paths, changes in purchased product categories, time series data, comparisons by category, etc.)," ​​and an appropriately designed SQL statement can be output. The visualization method can be determined by analyzing the semantic content of the request information, or a visualization method associated with a predetermined term can be selected. For example, in Figure 5 above, the request states, "I want to analyze the membership transition," and the word "transition" is included. Therefore, a line graph or bar graph can be determined as the visualization method, and this can be entered in the prompt.

[0060] That is, the prompt generation unit 103 can determine a visualization method according to the request information transmitted from the user terminal 400. By describing this visualization method in the prompt, the LLM 300 determines the "aggregation axis (time, category, etc.), which is the analysis method, and the aggregation unit (count, total, etc.)," ​​and changes the SQL to be extracted.

[0061] Next, the effects of the prompt generation device 100 of the present disclosure will be described. In the prompt generation device 100 of the present disclosure, the terminal access unit 101 functions as a receiving unit and receives request information for data processing from a user. In the present disclosure, data processing refers to processing for SQL searches, but naturally, other processing is also covered. Then, the acquisition unit 102 acquires related information related to the data processing based on the request information.

[0062] In the present disclosure, related information includes at least one of a data processing method, target information used in the data processing, and a method for analyzing result information obtained by the data processing. The data processing method is a search query, for example, an SQL query. The target information is a table describing the information used in the data processing. It is a table name expressed in the SQL query. The analysis method for the result information is a presentation method, a visualization method, such as the type of graph to be created.

[0063] In the prompt generation device 100, the prompt generation unit 103 generates a prompt based on such related information to instruct the generation AI to generate proposed content for data processing. The LLM access unit 104 sends it to the LLM 300 and obtains a response.

[0064] Furthermore, the acquisition unit 102 acquires related information related to data processing stored in the in-house DB 200 (corresponding to an external DB) based on the request information. That is, the acquisition unit 102 acquires related information similar to (i.e., related to) the request information based on the similarity between the vectors of the related information and the vectors of the request information. The in-house DB 200 stores, for example, past SQL queries. The acquisition unit 102 acquires information related to data processing, such as past SQL queries, and the prompt generation unit 103 generates a prompt based on the information. The prompt generation unit 103 sends the prompt to the LLM 300, thereby allowing the LLM 300 to propose a data processing method according to the user's request information.

[0065] The prompt also includes instructions to have the generation AI respond with a suggestion that satisfies the requested information and the reason for the suggestion.

[0066] This allows the user to obtain an answer that is satisfactory to the user.

[0067] The prompt generation device 100 of the present disclosure further includes an employee DB (user DB) that stores related information and processing frequency for each user (e.g., the number of times an SQL query has been written, information about DB operation experience, etc.). The prompt generation unit 103 generates prompts based on related information (e.g., search queries) of users with high processing frequencies.

[0068] This makes it easier to suggest similar search queries for users who frequently write search queries, resulting in better suggestions.

[0069] The device and method of the present disclosure have the following configuration.

[0070] [1] An apparatus comprising: a receiving unit that receives request information for data processing from a user; an acquisition unit that acquires related information related to the data processing based on the request information; and a generation unit that generates a prompt to instruct a generation AI to generate suggested content for the data processing based on the related information.

[0071] [2] The device according to [1], wherein the acquisition unit acquires related information related to data processing stored in an external DB based on the request information.

[0072] [3] The device according to [2], wherein the acquisition unit acquires the related information vector and the request information vector based on a similarity between the related information vector and the request information vector.

[0073] [4] The device according to any one of [1] to [3], wherein the related information includes at least one of a method of the data processing, target information used in the data processing, or a method of analyzing result information obtained by the data processing.

[0074] [5] The device according to any one of [1] to [4], wherein the prompt includes an instruction to have the generation AI respond with a suggestion that satisfies the requested information and a reason for the suggestion.

[0075] [6] The apparatus according to [4], wherein the data processing method is a search query.

[0076] [7] The device according to [4], wherein the target information is a table describing information used in the data processing. [8] The device according to [4], wherein the analysis method of the result information is a presentation method.

[0077] [9] The device according to any one of [1] to [8], further comprising a user DB that stores the related information and processing frequency for each user, and the prompt generation unit generates a prompt based on the related information of the user with a high processing frequency.

[0078]

[10] A method comprising: a receiving step of receiving request information for data processing from a user; an acquisition step of acquiring related information related to the data processing based on the request information; and a generation step of generating a prompt to cause a generation AI to propose the data processing based on the related information.

[0079] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (e.g., via wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0080] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0081] For example, the prompt generation device 100 according to an embodiment of the present disclosure may function as a computer that performs processing of the prompt generation method of the present disclosure. Fig. 8 is a diagram illustrating an example of the hardware configuration of the prompt generation device 100 according to an embodiment of the present disclosure. The prompt generation device 100 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage device 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0082] In the following description, the term "device" may be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the prompt generation device 100 may be configured to include one or more of the devices shown in the figures, or may be configured to exclude some of the devices.

[0083] Each function in the prompt generating device 100 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0084] The processor 1001, for example, runs an operating system to control the entire computer. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, the above-mentioned prompt generation unit 103 may be realized by the processor 1001.

[0085] The processor 1001 also loads programs (program code), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the prompt generation unit 103 may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be used for other functional blocks. While the above-described various processes have been described as being executed by a single processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0086] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be referred to as a register, a cache, a main memory (primary storage device), etc. The memory 1002 may store executable programs (program codes), software modules, etc. for implementing a prompt generation method according to one embodiment of the present disclosure.

[0087] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0088] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned terminal access unit 101, acquisition unit 102, LLM access unit 104, etc. may be realized by the communication device 1004. The communication device 1004 may be implemented with a transmitter and a receiver that are physically or logically separated from each other.

[0089] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0090] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0091] Furthermore, prompt generation device 100 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by such hardware. For example, processor 1001 may be implemented using at least one of these pieces of hardware.

[0092] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0093] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0094] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0095] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0096] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0097] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0098] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0099] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0100] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0101] Note that terms described in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0102] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by an index.

[0103] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0104] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," and the like may be used interchangeably.

[0105] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0106] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0107] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0108] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0109] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0110] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0111] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0112] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0113] 100 prompt generation device, 200 in-house DB, 300 LLM, 400 user terminal, 101 terminal access unit, 102 acquisition unit, 103 prompt generation unit, 104 LLM access unit.

Claims

1. An apparatus comprising: a reception unit that receives request information for data processing from a user; an acquisition unit that acquires related information related to the data processing based on the request information; and a generation unit that generates a prompt to instruct a generation AI to generate suggested content for the data processing based on the related information.

2. The device according to claim 1, wherein the acquisition unit acquires related information relating to data processing stored in an external DB based on the request information.

3. The device according to claim 2, wherein the acquisition unit acquires the related information based on the degree of similarity between the vector of the related information and the vector of the request information.

4. The device according to claim 1, wherein the related information includes at least one of a method of the data processing, target information used in the data processing, or a method of analyzing result information obtained by the data processing.

5. The device of claim 1, wherein the prompt includes an instruction to have the generating AI respond with a suggestion to satisfy the requested information and a reason for the suggestion.

6. The apparatus of claim 4, wherein the method of data processing is a search query.

7. The device according to claim 4, wherein the target information is a table describing information used in the data processing.

8. The device of claim 4, wherein the method of analyzing the result information is a presentation method.

9. The device according to claim 1, further comprising a user DB that stores processing frequencies for each user, wherein the generating unit generates prompts based on information related to the users with high processing frequencies.

10. A method comprising: a receiving step for receiving request information for data processing from a user; an acquisition step for acquiring related information related to the data processing based on the request information; and a generation step for generating a prompt to cause a generation AI to propose the data processing based on the related information.

Images