Information processing method, program, and information processing system

Abstract

An information processing method, a program, and an information processing system for effectively obtaining a desired processing result are provided. [Solution] In an information processing system including a server as an information processing device and a plurality of user terminals connected to each other so as to be able to communicate with each other via a network, an information processing method executed by the server includes a processing instruction acquisition step of acquiring processing instruction information that instructs processing, a user information acquisition step of acquiring user information about a user who inputs the processing instruction information, a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information, and a processing step of executing processing corresponding to the processing instruction information based on the usage information, using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input.

Description

[Technical field]

[0001] The present invention relates to an information processing method, a program, and an information processing system. [Background technology]

[0002] Patent Literature 1 discloses an information processing device that improves the prediction accuracy of a machine learning model. In this information processing device, in training a reliability judgment model for evaluating the reliability of a correct answer label used in training a machine learning model, additional information including the freshness of the information is collected and used to more accurately evaluate the reliability of the correct answer label. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-69344 A Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, it is known that the processing results output by a trained learned model change depending on the data (usage information) input to the trained learned model, but there is a need to control the usage information so that the desired processing results are obtained based on processing instruction information that instructs the processing.

[0005] An object of the present invention is to effectively obtain a desired processing result. [Means for solving the problem]

[0006] According to an information processing method of one embodiment, the information processing method is executed by an information processing device, and includes: a processing instruction acquisition step of acquiring processing instruction information that instructs processing; a user information acquisition step of acquiring user information related to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information; and a processing step of executing processing corresponding to the processing instruction information based on the usage information, using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input.

[0007] According to a program of one embodiment, an information processing device is caused to execute an information processing method including: a processing instruction acquisition step of acquiring processing instruction information that instructs processing; a user information acquisition step of acquiring user information related to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information; and a processing step of executing processing corresponding to the processing instruction information based on the usage information, using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input.

[0008] According to one embodiment, the information processing system is executed by an information processing device, and includes a processing instruction acquisition unit that acquires processing instruction information that instructs processing, a user information acquisition unit that acquires user information related to a user who inputs the processing instruction information, a usage information acquisition unit that acquires usage information required for processing from an information source based on the processing instruction information and the user information, and a processing unit that executes processing corresponding to the processing instruction information based on the usage information by utilizing a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input. Effect of the Invention

[0009] According to one embodiment, a desired processing result can be obtained efficiently. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of a configuration of an information processing system according to a first embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of a server according to the first embodiment. [Diagram 3] FIG. 2 is a diagram illustrating an example of a functional configuration of a server according to the first embodiment. [Figure 4] FIG. 2 is a diagram showing an example of a processing flow of the information processing system according to the first embodiment. [Diagram 5] FIG. 11 is a diagram showing an example of a screen display in an information processing system according to the second embodiment. [Figure 6] FIG. 11 is a diagram illustrating an example of a functional configuration of a server according to the second embodiment. [Figure 7] FIG. 11 is a diagram illustrating an example of a processing flow of an information processing system according to a second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] First embodiment A first embodiment of an information processing system according to the present invention will be described below with reference to Figures 1 to 4. In each drawing, the same or equivalent components and parts are given the same reference numerals. Also, the dimensional ratios in the drawings are exaggerated for the convenience of explanation and may differ from the actual ratios.

[0012] (System Overview) First, an overview of the information processing system 10 according to this embodiment will be described. The information processing system 10 according to this embodiment is a system in which, when a prompt (see Pt in FIG. 5) as processing instruction information instructing processing is input to a trained model such as a large-scale language model, the trained model generates a processing result (see Ot in FIG. 5) corresponding to the prompt and outputs the processing result. That is, when a prompt is input by a user U, the information processing system 10 causes the trained model to acquire usage information required for processing from a predetermined information source based on the prompt, and performs processing while referring to the usage information to generate a processing result. Details will be described later.

[0013] (System Configuration) Fig. 1 is a diagram showing an example of a configuration of an information processing system 10 according to the present embodiment. As shown in Fig. 1, the information processing system 10 according to the present embodiment includes a server 12 as an information processing device and a plurality of user terminals 14, which are communicably connected to each other via a network N. The network N is, for example, a wired LAN (Local Area Network), a wireless LAN, the Internet, a public line network, a mobile data communication network, or a combination of these.

[0014] The user terminal 14 is an example of an information processing device on which the user U, the manager M, and the operator OP perform operations for inputting and displaying various information. The user terminal 14 may be a PC (Personal Computer), a smartphone, a tablet terminal, a server device, a microcomputer, a wearable device, or a combination of these. As an example, the user U is a general employee, the manager M is a managerial staff member, and the operator OP is a service provider that provides services related to the information processing system 10. The user U, the manager M, and the operator OP each have account information as user information linked with predetermined authority information, and by logging in to the information processing system 10 using the account information, the information processing system 10 can obtain information on the account information and the usage status of the account information (details will be described later).

[0015] The server 12 is an example of an information processing device that acquires information input from the user terminal 14, processes the information, and outputs the results. The server 12 may be a PC (Personal Computer), a smartphone, a tablet terminal, a server device, a microcomputer, or a combination of these. The specific configuration and operation of the server 12 will be described later.

[0016] (Hardware configuration) 2 is a block diagram showing a hardware configuration of the server 12. The server 12 includes a processor 120, a memory 122, a storage 124, a communication I / F 126, an input / output I / F 128, an output device 130, an input device 132, and a drive device 134, which are communicatively connected to each other via a bus B.

[0017] The processor 120 controls each component of the server 12 and realizes the functions of the server 12 by expanding various programs stored in the storage 124 into the memory 122 and executing them. The programs executed by the processor 120 include, but are not limited to, an OS (Operating System) and a program 220 described below. The execution of these programs by the processor 120 realizes a part of the state visualization method according to this embodiment. The processor 120 is, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), a DSP (Digital Signal Processor), or a combination of these.

[0018] The memory 122 is, for example, a Read Only Memory (ROM), a Random Access Memory (RAM), or a combination thereof. The ROM is, for example, a Programmable ROM (PROM), an Erasable Programmable ROM (EPROM), an Electrically Erasable Programmable ROM (EEPROM), or a combination thereof. The RAM is, for example, a Dynamic RAM (DRAM), a Static RAM (SRAM), a Magnetoresistive RAM (MRAM), or a combination thereof.

[0019] The storage 124 stores an OS, various programs described below, and various data. The storage 124 is, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), a storage class memory (SCM), or a combination of these.

[0020] The communication I / F 126 is an interface for connecting the server 12 to external devices including the user terminal 14 and the image capturing device 16 via the network N and for controlling communication. The communication I / F 126 is, for example, an adapter conforming to Bluetooth (registered trademark), Wi-Fi (registered trademark), ZigBee (registered trademark), Ethernet (registered trademark), or optical communication (for example, Fibre Channel), but is not limited thereto.

[0021] The input / output I / F 128 is an interface for connecting an input device 132 and an output device 130 to the server 12. The input device 132 is, for example, a mouse, a keyboard, a touch panel, a microphone, a scanner, a camera, various sensors, an operation button, or a combination of these. The output device 130 is, for example, a display, a projector, a printer, a speaker, a vibrator, or a combination of these.

[0022] The drive device 134 reads and writes data from and to the disk medium 136. The drive device 134 is, for example, a magnetic disk drive, an optical disk drive, a magneto-optical disk drive, or a combination of these. The disk medium 109 is, for example, a compact disc (CD), a digital versatile disc (DVD), a floppy disk (FD), a magneto-optical disk (MO), a Blu-ray (registered trademark) disc (BD), or a combination of these.

[0023] In this embodiment, the program may be written into memory 122 or storage 124 during the manufacturing stage of server 12, or may be provided to server 12 via network N, or may be provided to server 12 via a non-transitory computer-readable recording medium such as disk media 136.

[0024] Moreover, the hardware configuration of the user terminal 14 is substantially the same as the hardware configuration of the server 12 described above, and therefore a detailed description thereof will be omitted.

[0025] (Functional configuration) Next, the functional configuration of the server 12 will be described. Fig. 3 is a diagram showing an example of the functional configuration of the server 12. When executing various programs, the server 12 realizes various functions using the above-mentioned hardware resources. The server 12 has a communication unit 20, a storage unit 22, and a control unit 24 as the functional configuration realized by the server 12. Each functional configuration is realized by the processor 120 reading and executing a program 220 stored in the memory 122 or the storage 124.

[0026] The communication unit 20 is realized by the communication I / F 126. The communication unit 20 transmits and receives information to and from the user terminal 14 via the network N. The communication unit 20 receives information input from the user terminal 14. In addition, the communication unit 20 transmits information to the user terminal 14, and receives a request by the user U from the user terminal 14.

[0027] The storage unit 22 is realized by the memory 122 and the storage 124. The storage unit 22 stores a program 220, a trained model 222, a prompt DB 224, an LV1 usage information DB 226, an LV2 usage information DB 228, a processing result DB 230, and a user information DB 232.

[0028] The trained model 222 is composed of at least one trained machine learning model. As an example, the trained model 222 is a large-scale language model that is trained with a large amount of text data from Internet articles, books, websites, etc., and when text data is input as a prompt, the trained model 222 generates and outputs text data corresponding to the prompt. The large-scale language model in the trained model 222 of this embodiment determines the direction of the output result based on the content of the prompt. Note that this output result refers to an expected output result.

[0029] Here, the direction indicates the nature or characteristic of the output result (hereinafter, simply referred to as "nature"). The trained model 222 judges the following two properties as an example. That is, the first property is a "strict property based on an information source" property (hereinafter, simply referred to as "strict"), and the output result according to this property is based on clear information extracted or derived from a specific information source. On the other hand, the second property is a "broad and creative" property (hereinafter, simply referred to as "creativity"), and the output result according to this property is not limited to a specific information source, but draws insight from a wide range of information and provides a new perspective or interpretation. The trained model 222 judges the degree of "strictness" and "creativity", which are opposing properties of the output result, by judging the direction. That is, when the prompt instructs "Please answer based on the contents of the previous management meeting", the trained model 222 judges that an output result of a strict nature based on an information source such as the minutes of the previous management meeting is required, and judges the direction of the output result to be "strict". On the other hand, if the prompt instructs "Please answer taking into consideration recent economic and technological trends," the trained model 222 determines that an output result of a creative nature that is not limited to a specific information source is required, and determines the direction of the output result as "creativity." Note that the above-mentioned determination of "strict" and "creativity" is merely an example, and it goes without saying that the direction of other prompts is appropriately determined based on the content of the prompts. In addition, the direction may be determined based on other properties. Furthermore, instead of determining only the presence or absence of each property, the degree to which the property is required may be determined numerically, such as "strictness 80, creativity 20." Furthermore, the output result may be divided into multiple parts, and the property may be determined for each part. As a specific example, the output result may be divided into a first part that provides a fact-based answer and a second part that provides consideration of the above part, and the first part may be determined as "strict" and the second part as "creativity."

[0030] Furthermore, the trained model 222 determines candidates for the usage information required for the process based on the prompt. Specifically, the trained model 222 evaluates the characteristics of the information required for the process based on the prompt (whether the information is based on common sense or on specialized knowledge, etc.), and when it is determined that the characteristics require acquisition from an external information source, etc., the trained model 222 searches the information source and determines the information required for the process as a candidate for the usage information. Note that the information source may be the storage unit 22 in which various information is stored, or may be an external information source such as a book or information on the Internet. Furthermore, the usage information is various information including text information stored in these information sources, and will be described in detail later.

[0031] Furthermore, the trained model 222 generates at least a part of the index information of the candidate usage information. At least a part of this index information is, for example, information freshness information, and this freshness information changes depending on the content of the usage information and the processing content. That is, the freshness information is, in other words, the "value of information on the time axis", and is roughly divided into a pattern in which the value decreases over time (hereinafter simply referred to as a "decreasing pattern"), a pattern in which the value is constant over time (hereinafter simply referred to as a "constant pattern"), and a pattern in which the value increases over time (hereinafter simply referred to as an "increasing pattern"). Each pattern will be explained below. Note that the "value of information" here indicates how much the information contributes to the processing result desired by the user U, and when the value of information is high, it contributes highly to the output of the desired processing result.

[0032] Decreasing patterns include a pattern in which value decreases at a substantially constant rate and a pattern in which value decreases suddenly. Information with a pattern in which value decreases at a substantially constant rate specifically corresponds to general news information, product information for which new models are periodically released, and information on the results of events, etc. These types of information are time-limited information whose value gradually decreases over time. Information with a pattern in which value decreases suddenly specifically corresponds to stock market information, weather forecasts, and news information related to events with rapid changes in circumstances, such as accidents and disasters.

[0033] Examples of information with a fixed pattern include information about historical facts and scientific principles, whose value changes relatively little over time.

[0034] The rising pattern includes a pattern in which the value rises at an approximately constant rate and a pattern in which the value rises rapidly. Information in a pattern in which the value rises at an approximately constant rate is information such as historical data, classical literature, products and old books that are produced in limited quantities, etc., which may not be recognized as being very valuable at first, but whose importance and value are re-recognized over time and whose value rises. Information in a pattern in which the value rises rapidly is information such as information on past artworks and collectibles, information on predictive analyses and predictions, information on specific technologies and theories, and information on vintage products, which did not attract much attention at first, but whose value rises rapidly as a result of attracting attention due to events such as the death of the creator or the later success of past predictions. Note that the specific information of each pattern is not limited to those described above.

[0035] Furthermore, the above-mentioned three patterns are merely examples, and other patterns may be included, such as a pattern in which the value fluctuates periodically, such as information about a periodic event. Furthermore, patterns that combine the above-mentioned patterns may be included. That is, examples include information whose value increases remains constant or decreases after a predetermined timing, information whose value was constant increases or decreases after a predetermined timing, and information whose value decreases remains constant or increases after a predetermined timing. In other words, the trained model 222 can also be expressed as "evaluating usage information whose value changes depending on the acquired prompt, using an index called freshness information."

[0036] The prompt DB 224 is a database that stores information indicating a prompt input by the user U. The prompt is written in a natural language, and the prompt DB 224 stores, as an example, a prompt and account information at the time the prompt was input in association with each other.

[0037] The LV1 usage information DB226 and the LV2 usage information DB228 as storage units are databases in which usage information required for processing is stored. The usage information stored in the LV1 usage information DB226 and the LV2 usage information DB228 is information acquired from information sources such as books, the Internet, and the storage unit 22 as described above, and includes text information, image information, and the like. The usage information stored in the LV1 usage information DB226 and the LV2 usage information DB228 includes unstructured data and structured data. Here, "unstructured data" refers to data that is stored without being structured in a predefined manner. On the other hand, "structured data" refers to data that exists in a fixed field in a file or record. "Structured data" also includes "semi-structured data." Although the storage units are the LV1 usage information DB226 and the LV2 usage information DB228, they are not limited to this, and may be three or more databases depending on a predetermined authority.

[0038] The LV1 usage information DB226 and the LV2 usage information DB228 are a plurality of storage units in which accessibility is set for each level of predetermined authority information. Specifically, the LV1 usage information DB226 is set to be accessible only when processing is performed by an account having authority information set for a general employee such as the user U. On the other hand, the LV2 usage information DB228 is set to be accessible only when processing is performed by an account having authority information set for a manager such as the manager M. In other words, when processing is performed by the user U, the usage information stored in the LV1 usage information DB226 can be acquired and processed, but the usage information stored in the LV2 usage information DB228 cannot be acquired. The same is true when processing is performed by another account (not shown) having the same authority information as the user U. The same is true when processing is performed by the manager M, but the usage information stored in the LV2 usage information DB228 can be acquired and processed, but the usage information stored in the LV1 usage information DB226 cannot be acquired. The same is true when processing is performed by another account (not shown) having the same authority information as the manager M. The operator OP can access both the LV1 usage information DB 226 and the LV2 usage information DB 228.

[0039] The processing result DB 230 is a database that stores information indicating the results of executing a process using usage information based on a prompt (also called the execution result of the process or the processing result). The processing result is various information including text information and image information, and the processing result DB 230 stores the processing result, the prompt, and the usage information in a linked manner.

[0040] The user information DB 232 is a database that stores various information of the user U and the manager M who have registered to use the information processing system 10. Examples of the various information stored include user identification information for unique identification, user preference information such as interest information, user history information including past usage status, and user attribute information such as occupation, position, and authority.

[0041] The control unit 24 is realized by the processor 120 reading and executing the program 220 from the memory 122 (see FIG. 2) and cooperating with other hardware configurations. The control unit 24 includes an information acquisition unit 242, an information processing unit 244, a determination result acquisition unit 246, and an output unit 248.

[0042] The information acquisition unit 242 acquires various information required for processing in the information processing unit 244. Specifically, it acquires usage information including unstructured data required for processing from at least one of the prompt DB 224, the user terminal 14, and an external data source. The information acquisition unit 242 also acquires index information of the usage information (details will be described later). Furthermore, the information acquisition unit 242 acquires account information logged in and operating the information processing system 10. In other words, the information acquisition unit 242 corresponds to the processing instruction acquisition unit, user information acquisition unit, and usage information acquisition unit recited in claim 6.

[0043] The information processing unit 244 determines the directionality of the output result based on the contents of the prompt, selects a storage unit to acquire usage information from based on account information, acquires and generates index information in the usage information, determines the degree of agreement between the directionality of the output result and the usage information, and executes processing based on the prompt. The directionality of the output result is determined by using the trained model 222 to determine the nature of the output result expected in response to the prompt input by the user U, as described above.

[0044] The information processing unit 244 judges the predetermined authority information in the acquired account information and selects a storage unit accessible by the account. That is, when the authority information in the acquired account information is set to a general employee such as a user U, the information processing unit 244 selects the LV1 usage information DB 226 accessible by the authority information as an information source. Similarly, when the authority information in the acquired account information is set to a managerial position such as a manager M, the information processing unit 244 selects the LV2 usage information DB 228 accessible by the authority information as an information source. Moreover, when the authority information in the acquired account information is set to a system administrator such as an operator OP, the information processing unit 244 selects the LV1 usage information DB 226 and the LV2 usage information DB 228 accessible by the authority information as information sources. Then, the information processing unit 244 acquires the usage information from the selected storage unit.

[0045] The information processing unit 244 generates freshness information from the index information of the usage information by using the trained model 222 as described above. The information processing unit 244 is also able to use metadata and log data as index information of the usage information acquired by the information acquisition unit 242. The metadata is information that describes the characteristics and attributes of data, such as the content, source, creator, and timestamp of the usage information. The log data is information that is recorded along with the time generated each time a certain operation is performed, such as the date and time when the usage information was recorded or accessed, the executor who performed it, request and response information, and error information. These metadata and log data are linked to the usage information in advance and are able to be used. The information processing unit 244 uses at least one of the freshness information, metadata, and log data, which are index information, to determine whether the usage information matches the directionality of the output result (details will be described later).

[0046] The information processing unit 244 judges the degree of match between the directionality of the output result and the usage information. That is, the degree of match is judged based on the index information of the usage information as to whether the candidate usage information matches the prompt-based processing with respect to the directionality. As a specific example, when the directionality is judged to be "strict", the degree of match of the usage information is judged using the index information from the viewpoints of whether the candidate usage information is accurate, complete, up-to-date, reliable, clear, traceable, consistent, etc.

[0047] In other words, accuracy is determined based on criteria such as, for example, whether the update date and time of the usage information in the metadata is recent and whether that date and time has been recorded accurately, whether the registrant is a highly trustworthy person or institution, whether the metadata conforms to a predefined format or standard, whether error messages or abnormal patterns are searched for in the log data to see if there are any signs of a system abnormality, and if the freshness information shows an upward or downward pattern, the factors that led to that pattern are tracked to see if there are any inconsistencies with the content.

[0048] Completeness is determined based on criteria such as whether all necessary metadata elements are present, whether all log data that should be collected is present, and whether the freshness information contains important elements or bits that allow for an overall picture of changes in the information's value, such as the specific time periods when an increase or decrease in value occurred and the circumstances under which this occurred.

[0049] Whether or not information is up to date is determined based on criteria such as whether it reflects the latest information from the metadata, whether the log data is real-time or the latest data, whether the freshness information reflects the latest fluctuations in information value, and whether the information value is above a predetermined value.

[0050] Whether a data is highly reliable or not is determined based on criteria such as the reliability of the data source from the metadata, the evaluation results of the metadata generation process, whether the log data is collected using reliable means and there is no risk of tampering, and whether the freshness information shows little fluctuation in the information value in any pattern.

[0051] Whether or not something is clear is determined based on criteria such as whether the metadata is defined according to clear explanations and conventions, whether the log data is recorded in a consistent format and can be clearly interpreted, and whether the factors that cause the value of the information to decrease, remain constant, or increase can be clearly interpreted from the freshness information.

[0052] Whether or not something is traceable is determined based on criteria such as whether the source of the data is clearly indicated from the metadata and whether it is possible to query that source, whether each log entry can be traced over time from the log data (whether time information is included), and whether changes in each piece of information value can be traced over time from the freshness information, i.e., whether the time point at which the change occurred is clearly indicated.

[0053] Whether or not there is consistency is determined based on criteria such as whether there is consistency between metadata from the metadata and whether they comply with defined rules and standards, whether the log data is generated in a consistent format and pattern from the log data, whether the fluctuation in information value is consistent from the freshness information, etc. Note that the above-mentioned criteria are merely examples, and other criteria may be used for the determination.

[0054] When the degree of match between the directionality of the output result and the usage information is equal to or greater than a predetermined degree, the information processing unit 244 refers to the usage information and performs processing corresponding to the prompt by using the trained model 222. On the other hand, when the degree of match between the directionality of the output result and the usage information is not equal to or greater than a predetermined degree, the information processing unit 244 searches for the usage information again and acquires candidates for the usage information as a search result via the information acquisition unit 242.

[0055] The information processing unit 244 links the processing result corresponding to the prompt with the prompt and stores the prompt in the processing result DB 230. Furthermore, if the processing result includes usage information used in the processing, the information processing unit 244 links the usage information with the processing result DB 230 and stores the usage information.

[0056] Furthermore, when the information processing unit 244 acquires a new prompt, if the relevance between the prompt and a prompt already stored in the prompt DB 224 (hereinafter simply referred to as a "cache prompt") is equal to or greater than a predetermined criterion, the information processing unit 244 outputs the processing result associated with the cache prompt and stored in the processing result DB 230. As an example, the predetermined criterion is a criterion for determining that the new prompt and the prompt stored in the prompt DB 224 are similar to each other, and specifically, the new prompt and the prompt associated with the processing result stored in the prompt DB 224 are vector-converted and set to a value such as cosine similarity or Euclidean distance. Here, a value equal to or greater than the predetermined criterion indicates that the similarity is higher. Note that the predetermined criterion is not limited to the above-mentioned one, and may be other criteria such as word-based similarity or meaning-based similarity. Furthermore, the predetermined criterion may be a similarity determination not only in comparison between the new prompt and the prompt associated with the processing result stored in the prompt DB 224, but also in comparison between the new prompt and the processing result itself stored in the processing result DB 230. In this case, the similarity determination may be based on the similarity of values ​​obtained by vector conversion, word-based similarity, or meaning-based similarity, as described above. In an example of comparing a new prompt with the processing result stored in the processing result DB 230, a new prompt "How to upgrade your data plan" is compared with the content of a past processing result stored in the processing result DB 230, "You can change your plan from the account setting page," and it is determined that both are related to changing the data plan and have a high similarity. Furthermore, the predetermined criteria may be not only those that are determined to be similar, but also those that are determined to be connected, have something in common, have an influencing relationship, or other criteria.

[0057] The judgment result acquisition unit 246 acquires a judgment result as to whether the processing result is good or not in response to the prompt based on a predetermined criterion. That is, the judgment result acquisition unit 246 acquires feedback information as to whether the processing result input by the user U or the manager M is good or not based on a judgment as to whether the processing result by the user U or the manager M is a desired result in response to the prompt as a predetermined criterion. As an example, this feedback information is information in which the user U or the manager M who has seen the processing result presses the "Good" button 80 (see FIG. 5) through the user terminal 14 because the processing result is good, or information in which the user U or the manager M who has seen the processing result presses the "Bad" button 82 (see FIG. 5), not shown, through the user terminal 14 because the processing result is not good. Then, when the feedback information is good (the "Good" button 80 has been pressed), the judgment result acquisition unit 246 stores the processing result for which the feedback information has been performed in the processing result DB 230.

[0058] The output unit 248 controls the information processing unit 244 so that the processing result is output to the user terminal 14 .

[0059] (Processing Executed by Information Processing System 10) Next, the operation of the information processing system 10 will be described. Fig. 4 is a flow chart showing an example of the flow of processing by the information processing system 10. The processor 120 reads out the program 220 stored in the storage 124, expands it in the memory 122, and executes it, thereby performing processing. Although not shown, when the processor 120 receives operation information for terminating the operation of the information processing system 10, or information on the end of an operation from the user terminal 14 during the determination processing being executed (these are simply referred to as "termination operation"), the processor 120 ends the processing based on the program 220 being processed.

[0060] The processor 120 acquires account information logged into the information processing system 10 (step S100). Then, the processor 120 determines whether or not a prompt inputted by an account linked to the acquired account information has been acquired (step S102). If a prompt has not been inputted or a prompt cannot be acquired, such as during input (step S102: NO), the processor 120 repeats the process of step S102. On the other hand, if a prompt has been acquired (step S102: YES), the processor 120 determines the relevance with the cached prompt or the processing result stored in the processing result DB 230 (step S104), and determines whether or not the relevance is equal to or greater than a predetermined standard (step S106). If the relevance is equal to or greater than the predetermined standard (step S106: YES), the processor 120 acquires the processing result (including the processing result linked to the cached prompt) stored in the processing result DB 230 (step S108), and proceeds to step S124, which will be described later. On the other hand, if the relevance is equal to or lower than the predetermined standard (step S106: NO), the processor 120 determines the direction of the result to be output based on the contents of the prompt (step S110). Note that the above-mentioned step S100 corresponds to the "user information acquisition step" recited in claim 1, and step S102 corresponds to the "processing instruction acquisition step" recited in claim 1.

[0061] The processor 120 judges the predetermined authority information in the acquired account information, and selects a storage unit (at least one of the LV1 usage information DB 226 and the LV2 usage information DB 228) accessible by the account (step S112). Then, the processor 120 determines candidates of usage information required for the process based on the prompt in the accessible storage unit, and acquires index information of the candidate usage information (step S114). After that, the processor 120 judges whether or not there is usage information that matches the directionality determined in step S110 from the acquired index information (step S116). If there is no usage information with a matching directionality (step S116: NO), the processor 120 again searches and determines candidates of usage information required for the process based on the prompt in the accessible storage unit after excluding the usage information (step S118), and proceeds to step S114. On the other hand, if there is usage information with matching directionality (step S116: YES), the processor 120 acquires the usage information (step S120) and performs information processing using the usage information based on the prompt (step S122). The processor 120 then outputs the processing results (step S124). This output includes displaying on the display of the user terminal 14, executing processing on external tools, sensors, data sources, etc. Note that the above-mentioned step S120 corresponds to the "usage information acquisition step" recited in claim 1, and step S122 corresponds to the "processing step" recited in claim 1.

[0062] The processor 120 judges whether or not feedback information on the processing result has been acquired (step S126). If feedback information has not been acquired (step S126: NO), the processor 120 ends the processing. On the other hand, if feedback information has been acquired (step S126: YES), the processor 120 judges whether the feedback information is equal to or higher than a predetermined standard (in this embodiment, whether or not it is "Good") (step S128). If the feedback information is not equal to or higher than the predetermined standard, that is, if it is "Bad" (step S128: NO), the processor 120 ends the processing. On the other hand, if the feedback information is equal to or higher than the predetermined standard, that is, if it is "Good" (step S128: YES), the processor 120 stores the processing result in the processing result DB 230 (step S130) and ends the processing. Note that the above-mentioned step S126 corresponds to the "judgment result acquisition step" recited in claim 3, and step S130 corresponds to the "good result storage step" recited in claim 3.

[0063] (Effects of the First Embodiment) According to the information processing system 10 of the present embodiment, the following steps are executed: a processing instruction acquisition step of acquiring a prompt (processing instruction information) instructing a process, a user information acquisition step of acquiring user information on a user who inputs the processing instruction information, a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information, and a processing step of executing processing corresponding to the processing instruction information based on the usage information using a trained model that trains learning data including text data and outputs a processing result according to the processing instruction information when the processing instruction information is input. In other words, the usage information is acquired from an information source appropriate for the user based on the user information, and processing is performed based on a prompt using the usage information, so that the processing result desired by the user can be effectively obtained.

[0064] Further, the user information is associated with predetermined authority information set to at least one of a plurality of levels, and the information source is stored in one of a plurality of storage units in which accessibility is set for each level of the predetermined authority information. In the usage information acquisition step, the usage information is acquired from the information source in the storage unit accessible based on the predetermined authority information of the user information, so that the processing based on the prompt can be executed using appropriate usage information for each user's authority. That is, as an example, personal information and salary information of employees (hereinafter simply referred to as "confidential information") that should be disclosed only to users with managerial authority are stored in a storage unit (in this embodiment, the LV2 usage information DB228) that can be accessed only when processing by users with managerial authority. As a result, even if a user with general employee authority inputs a prompt inquiring about personal information and salary information of employees, the LV2 usage information DB228 cannot be accessed, so the confidential information stored in the LV2 usage information DB228 is not used as usage information to execute processing. That is, the storage unit in which the usage information is physically stored is used depending on the user's predetermined authority information, so that the learned model 222 can be prevented from acquiring unintended usage information and executing processing. This makes it possible to achieve both appropriate information management and effectively obtain a processing result desired by the user. In particular, in the case of processing using a large-scale language model, even if the content of the processing is specified in advance, there is a risk that processing different from the content specified by a prompt with a different expression will be performed (for example, secret information will be disclosed to an unauthorized user). However, by using a storage unit in which the usage information is physically stored according to the user's predetermined authority information, even if the processing result is different from the specified content (for example, information close to secret information will be disclosed to an unauthorized user), it becomes clear that the processing is hallucination using a large-scale language model. In other words, the cause of an unintended output result can be easily identified. This makes it easier to control the processing result.

[0065] The method further includes a judgment result acquisition step of acquiring a judgment result as to whether the execution result of the processing in the processing step is good or not for the prompt based on a predetermined criterion, and a good result storage step of storing the execution result of the processing determined to be good in a storage unit in association with the prompt, and outputs the execution result of the processing associated with the processing instruction information stored in the good result storage step when the association between the prompt acquired in the processing instruction acquisition step and the prompt stored in the good result storage step is equal to or greater than a predetermined criterion. This allows the processing result associated with the past prompt to be used when a past prompt and a newly acquired prompt are similar, thereby saving the calculation cost and time using the trained model 222. In addition, since the processing result judged to be good is stored in the storage unit, the processing result desired by the user can be effectively obtained.

[0066] Furthermore, since the information processing system 10 acquires the usage information based on the output directionality determined from the prompt in the usage information acquisition step, the information processing system 10 can execute the process corresponding to the prompt based on the appropriate usage information, thereby more effectively obtaining the desired processing result.

[0067] Furthermore, since the information processing system 10 uses the metadata of the usage information when acquiring the usage information in the usage information acquisition step, it is possible to efficiently and appropriately select the usage information D. This makes it possible to more effectively obtain the desired processing result.

[0068] Furthermore, since the information processing system 10 uses log data related to the usage information when acquiring the usage information in the usage information acquisition step, the usage information can be selected based on objective records, thereby making it possible to more effectively obtain the desired processing results.

[0069] Furthermore, in the usage information acquisition step, the information processing system 10 inputs information of the information source to the trained model 222 that has been trained to output index information of the information when the information is input, and uses the index information output from the trained model 222 when acquiring the usage information, so that the usage information can be appropriately selected even for data that does not have metadata or log data. This makes it possible to obtain the desired processing result more effectively.

[0070] Second Embodiment Next, an information processing system 40 according to a second embodiment of the present invention will be described with reference to Figures 5 to 7. The information processing system 40 according to the second embodiment has a basic configuration similar to that of the first embodiment, and is characterized in that it executes a separate processing step that executes a process based on a prompt by a means separate from the process in the processing step. Note that the same components as those in the first embodiment are given the same reference numerals and their description will be omitted.

[0071] (Functional configuration) The functional configuration of the server 50 as an information processing device in the information processing system 40 will be described. FIG. 6 is a diagram showing an example of the functional configuration of the server 50. When executing various programs, the server 50 realizes various functions using the same hardware resources as the server 12 of the first embodiment. The server 50 has a communication unit 20, a storage unit 60, and a control unit 70 as the functional configuration realized by the server 50. Each functional configuration is realized by the processor 120 reading and executing a program 600 stored in the memory 122 or the storage 124.

[0072] The control unit 70 is realized by the processor 120 reading and executing the program 600 from the memory 122 (see FIG. 2) and cooperating with other hardware configurations. The control unit 70 includes an information acquisition unit 242, an information processing unit 700 as a processing unit, a determination result acquisition unit 246, a separate processing unit 702, and an output unit 248.

[0073] Similar to the information processing unit 244 of the first embodiment, the information processing unit 700 determines the directionality of the output result based on the contents of the prompt, selects an accessible storage unit based on account information, obtains and generates index information for the usage information in the storage unit, determines the degree of match between the directionality of the output result and the usage information, and executes processing based on the prompt. The information processing unit 700 also obtains feedback information on the processing result corresponding to the prompt, and stores the processing result in the storage unit 60 by linking the prompt and the usage information, if any, based on the feedback information.

[0074] Further, the information processing unit 700 outputs the processing result (details will be described later) by the separate processing unit 702 depending on the situation. That is, as shown in FIG. 5, when the processing result by the information processing unit 700 is output to the user terminal 14 used by the user U or the manager M, if the user U or the manager M performs an operation of pressing the "HELP" button 84 on the processing result, the separate processing unit 702 is controlled to execute processing assuming that the user U or the manager M requests help / inquiry for the processing result. In this embodiment, the processing in the separate processing unit 702 is executed when feedback information of the user U or the manager M's intention to request help / inquiry for the processing result by the information processing unit 700 is acquired, but this is not limited to this, and feedback information indicating other intentions may also be acquired. Furthermore, the processing in the separate processing unit 702 may be executed not only based on the judgment of a person by the user U or the like, but also based on the judgment by the information processing device, such as when the information processing system 40 judges whether the processing result by the information processing unit 700 meets a predetermined quality standard and the quality standard is not met.

[0075] The separate processing unit 702 performs processing using the usage information based on a prompt by a means other than the information processing unit 700. In this embodiment, as an example, the separate processing unit 702 presents a prompt to the operator OP. Furthermore, the separate processing unit 702 obtains the result of the operator OP searching for usage information based on the prompt and performing processing using the usage information of the search result. Specifically, the separate processing unit 702 outputs a prompt to be processed so as to be displayed on the user terminal 14 of the operator OP, and controls access to, search for, and presentation of usage information in the LV1 usage information DB 226 and the LV2 usage information DB 228 for the operator OP to search for usage information taking user information into consideration. Furthermore, the separate processing unit 702 controls the user terminal 14 of the operator OP to accept input of the processing result by the operator OP, obtains the processing result input by the operator OP, and outputs the processing result instead of the processing result of the information processing unit 700 when the processing result is obtained. The case where the processing result by the operator OP is input corresponds to the "case where a predetermined condition is satisfied" described in claim 4. That is, the case where the processing result can be acquired by a means other than the information processing unit 700 corresponds to the case where the "predetermined condition is satisfied". In this embodiment, the separate processing unit 702 is assumed to perform processing by an operator OP, but is not limited to this, and may be processed by a user U or a manager M. That is, when feedback information of an intention to request help or an inquiry by a user U or a manager M is acquired, another user U or a manager M related to the feedback information may be identified from the account information, and the user U or the manager M may be prompted to perform processing or add usage information. That is, the separate processing unit 702 may be configured to perform processing by an appropriate person in charge. In addition to this, the processing of the separate processing unit 702 may use a trained model other than that of the information processing unit 700, or may use other methods such as a rule base.

[0076] (Processing Executed by Information Processing System 40) Next, the operation of the information processing system 40 will be described. Fig. 7 is a flowchart showing an example of the flow of processing by the information processing system 40. The processor 120 reads out the program 600 stored in the storage 124, expands it in the memory 122, and executes it, thereby performing processing. Note that the same processes as those in the first embodiment are denoted by the same reference numerals, and the description thereof will be omitted.

[0077] If feedback information is not acquired in the process of step S126 (step S126: NO), the processor 120 judges whether or not feedback information such as an inquiry is acquired (step S200). If feedback information such as an inquiry is not acquired (step S200: NO), the processor 120 ends the process. On the other hand, if feedback information such as an inquiry is acquired (step S200: YES), the processor 120 notifies the user terminal 14 of the operator OP with a prompt and requests that the operator OP perform processing (step S202).

[0078] The processor 120 judges whether or not the operator OP has input a processing result (step S204). If no input has been made (step S204: NO), the processor 120 shifts the process to step S202. On the other hand, if input has been made (step S204: YES), the processor 120 outputs the processing result (step S206). This output includes displaying the processing result on the display of the user terminal 14, executing processing on external tools, sensors, data sources, etc. Then, the processor 120 shifts the process to step S130. Note that the above-mentioned steps S202 and S204 correspond to the "separate processing step" recited in claim 4.

[0079] (Effects of the second embodiment) The information processing system 40 according to this embodiment has the same configuration as the first embodiment, except that a separate processing step is executed to execute a process based on a prompt by a means other than the process in the processing step, and therefore the same action and effect as the first embodiment can be obtained. In addition, the information processing system 40 includes a separate processing step in which a process based on a prompt is executed by a means other than the process in the processing step, and outputs the execution result of the separate processing step instead of the execution result of the processing step when the execution result of the process executed in the separate processing step satisfies a predetermined condition, so that when the processing result by the trained model 222 is not good, the processing result executed by the operator OP can be obtained. This makes it possible to obtain a more desired processing result effectively.

[0080] Furthermore, the processing result of the separate processing step is stored in the storage unit 22 after the result is output. In other words, if there is a high correlation between a prompt in the processing result of the separate processing step and the new prompt when it is acquired, the processing result of the separate processing step is preferentially output. This gives priority to the processing result of the separate processing step, the quality of which is guaranteed to a certain extent (in this embodiment, the quality of the processing result is guaranteed by the processing by the operator OP), so that a more desired processing result can be obtained effectively.

[0081] In the second embodiment described above, the "HELP" button 84 is displayed alongside the "Good" button 80 and the "Bad" button 82 for transmitting feedback information (see FIG. 5), but the present invention is not limited to this, and the "HELP" button 84 may be displayed only after the "Bad" button 82 is pressed. In this case, it is possible to prevent the user U from easily pressing the "HELP" button 84, and therefore it is possible to effectively obtain a desired processing result while preventing an increase in the load due to an increase in the execution of separate processing steps.

[0082] In the first and second embodiments described above, at least one of metadata, log data, and freshness information is used as the index information of the usage information DB226. However, other data such as annotation data may be used as the index information. The trained model 222 is configured to generate freshness information of the usage information DB226 from the contents of the usage information DB226. However, the freshness information may be generated using not only the contents of the usage information DB226 but also information such as metadata and log data. The trained model 222 may generate not only freshness information as index information but also information for determining information necessary for processing, such as a summary of the contents of the usage information DB226. The information processing systems 10 and 40 may be configured to determine these pieces of information using at least one of the rule base and the trained model 222 and acquire the usage information DB226 necessary for processing from the information source.

[0083] Furthermore, the freshness information of the usage information DB226 is generated by the trained model 222 that performs processing based on a prompt, but the present invention is not limited to this. The index information including the freshness information may be generated by a trained model other than the trained model 222 that performs processing based on a prompt. In this configuration, a trained model optimized for generating index information can be used, so that the usage information DB226 can be acquired with higher accuracy. The trained model that outputs the freshness information is, for example, a model that has been machine-learned using a data set that uses at least one of time-stamped information, origin information, update frequency information, user reaction information, and topic / field property information. Here, the time-stamped information is information to which a timestamp indicating when the target information was generated is added, and makes it possible to directly evaluate the freshness of the information. The origin information is information indicating where the target information came from, and, for example, a newspaper article is most fresh on the day of its publication, and its freshness decreases over time, whereas historical data and cultural data have higher information value as time passes, making it possible to evaluate the freshness of information. The update frequency information indicates how frequently the target information is updated, and enables evaluation of the freshness of the information, for example, that the more frequently the information is updated, the higher the value of the information. The user reaction information indicates how the target information was reacted to, for example, the number of clicks, display time, number of shares, and comments, and enables evaluation of the freshness of the information, for example, that information with a large amount of reaction information has high information value. The topic / field characteristic information enables evaluation of the freshness of the information by taking into account the background of each field, for example, that in the field of science and technology, information is updated every time new research or publication is made, but in fields such as literature and history, information is not updated so frequently.

[0084] Furthermore, in the processing performed by processor 120 in step S116 in the first and second embodiments described above, the configuration is such that it is determined whether or not there is usage information that matches the directionality determined in step S110 from the acquired index information, but this is not limited to this. It is also possible to calculate the degree of match with the directionality in the acquired index information, and acquire usage information corresponding to the index information in order of highest degree of match and use it for processing, or it is also possible to combine the determination based on this degree of match with the determination of the presence or absence of the usage information described above.

[0085] In addition, although the configuration is such that feedback information is acquired as a judgment result as to whether or not the processing result is good, the configuration may be such that it is judged whether or not the processing result is good when the user U or the manager M inputs a prompt to further confirm the contents after the processing result is output. In other words, if a prompt to further confirm the contents is input after the processing result is output, it can generally be judged that the processing result is not good. In addition, the configuration may be such that it uses information that judges the quality of the processing result using a trained model.

[0086] (Variation 1) Furthermore, if the above-mentioned information processing system is viewed from a different perspective, the problem (purpose) that the information processing system of this embodiment is trying to solve can also be viewed as "being able to use processing results from a different perspective for the processing results using a trained model."

[0087] If the problem is understood as above, the invention as a means for solving the problem is, for example, as follows. "An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; a separate processing step of executing a process based on the processing instruction information by a means separate from the process in the processing step; When the execution result of the processing executed in the separate processing step satisfies a predetermined condition, the execution result of the processing in the separate processing step is output instead of the execution result of the processing in the processing step. Information processing methods.”

[0088] According to the above configuration, the system includes a separate processing step in which a process based on a prompt is executed by a means other than the process in the processing step, and when the execution result of the process executed in the separate processing step satisfies a predetermined condition, the execution result of the process in the separate processing step is output instead of the execution result of the process in the processing step, so that when the processing result by the trained model is not satisfactory, a processing result by another means such as a processing result executed by an operator OP can be obtained. This makes it possible to use a processing result from a different perspective for the processing result by the trained model, thereby making it possible to provide a highly reliable service.

[0089] (Variation 2) Furthermore, if we look at the above-mentioned information processing system from a different perspective, the problem (purpose) that the information processing system of this embodiment aims to solve can also be seen as "being able to encourage the updating or addition of usage information."

[0090] If the problem is understood as above, the invention as a means for solving the problem is, for example, as follows. "An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; a determination result acquisition step of acquiring a determination result as to whether or not the execution result of the processing in the processing step is satisfactory with respect to the processing instruction information based on a predetermined criterion; an update notification step of notifying at least one of updating and adding the usage information used in the processing step when the judgment result is not good; "An information processing method including the following."

[0091] According to the above configuration, if the judgment result is not good in the judgment result acquisition step, at least one of updating and adding the usage information used when the processing result is calculated is notified, so that the user, operator, etc. can update or add the usage information based on the notification. In other words, it is possible to continuously improve the usage information. Therefore, it is possible to continuously improve the processing result.

[0092] (Variation 3) Furthermore, if we look at the above-mentioned information processing system from a different perspective, the problem (purpose) that the information processing system of this embodiment aims to solve can also be seen as "achieving both savings in computational costs and time and effectively obtaining processing results that the user desires."

[0093] If the problem is understood as above, the invention as a means for solving the problem is, for example, as follows. "An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; a determination result acquisition step of acquiring a determination result as to whether or not the execution result of the processing in the processing step is satisfactory with respect to the processing instruction information based on a predetermined criterion; a good result storage step of storing the execution result of the process determined to be good in a storage unit in association with the processing instruction information; a cache control step of controlling a timing for deleting a result of the processing from the storage unit based on at least one of the contents and characteristics of the execution result of the processing stored in the storage unit; Including, When the correlation between the processing instruction information acquired in the processing instruction acquisition step and the processing instruction information stored in the good result storage step is equal to or greater than a predetermined standard, an execution result of the processing associated with the processing instruction information stored in the good result storage step is output. Information processing methods.”

[0094] According to the above configuration, the timing of deleting the processing result stored in the storage unit as a result of the judgment being good is controlled based on at least one of the contents and characteristics in the cache control step, so that processing results with outdated contents or processing results having characteristics biased toward a specific bias can be deleted from the storage unit at an appropriate timing. As a result, if a past prompt and a newly acquired prompt are similar, the processing result linked to the past prompt can be used, so that the calculation cost and time using a trained model can be saved. On the other hand, the output of an inappropriate processing result can be suppressed, so that the processing result desired by the user can be effectively obtained. The "characteristics" refer to the consistency, accuracy, naturalness (whether the use of grammar and vocabulary is natural and easy to understand), diversity, creativity, the presence or absence of bias, etc. of the processing result.

[0095] (Variation 4) Furthermore, if we look at the above-mentioned information processing system from a different perspective, the problem (purpose) that the information processing system of this embodiment aims to solve can also be seen as "improving the user's reliability of processing results."

[0096] If the problem is understood as above, the invention as a means for solving the problem is, for example, as follows. "An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a past processing result acquisition step of searching for past processing results stored in a storage unit based on the processing instruction information and acquiring related past processing results; a processing step of executing a process corresponding to the processing instruction information based on a past processing result in the past processing result acquisition step, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; "An information processing method including the following."

[0097] According to the above configuration, a process corresponding to the processing instruction information is executed based on a past processing result. That is, a new prompt is generated from the usage information and the past processing result based on the processing instruction information, and the prompt is input to the trained model to execute the process. This allows the processing result to be in accordance with a context related to the past processing result. This allows the processing result to be consistent, thereby improving the reliability of the user's processing result. Note that the above configuration is configured to execute a process corresponding to the processing instruction information based on the past processing result acquired in the past processing result acquisition step, but is not limited to this, and may also be configured to use the usage information acquired in the "usage information acquisition step" described in claim 1.

[0098] <Additional Notes> The present embodiment includes the following disclosure.

[0099] (Appendix 1) An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a user information acquisition step of acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information; A processing step of executing a process corresponding to the processing instruction information based on the usage information using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; An information processing method comprising:

[0100] (Appendix 2) The user information is associated with predetermined authority information that is set to at least one of a plurality of levels, the information source is stored in one of a plurality of storage units in which access permission is set for each level of the predetermined authority information, In the usage information acquisition step, the usage information is acquired from the information source in the storage unit that is accessible based on the predetermined authority information of the user information. 2. An information processing method as described in Appendix 1.

[0101] (Appendix 3) a determination result acquisition step of acquiring a determination result as to whether or not the execution result of the processing in the processing step is satisfactory with respect to the processing instruction information based on a predetermined criterion; a good result storage step of storing the execution result of the process determined to be good in a storage unit in association with the processing instruction information; Including, When the correlation between the processing instruction information acquired in the processing instruction acquisition step and the processing instruction information stored in the good result storage step is equal to or greater than a predetermined standard, an execution result of the processing associated with the processing instruction information stored in the good result storage step is output. 10. The information processing method according to claim 1 or 2.

[0102] (Appendix 4) a separate processing step of executing a process based on the processing instruction information by a means separate from the process in the processing step, When the execution result of the processing executed in the separate processing step satisfies a predetermined condition, the execution result of the processing in the separate processing step is output instead of the execution result of the processing in the processing step. An information processing method according to any one of Supplementary Note 1 to Supplementary Note 3.

[0103] (Appendix 5) In the information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a user information acquisition step of acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information and the user information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; A program for executing an information processing method including the steps of:

[0104] (Appendix 6) An information processing system executed by an information processing device, a processing instruction acquisition unit that acquires processing instruction information that instructs processing; a user information acquiring unit for acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition unit that acquires usage information required for processing from an information source based on the processing instruction information and the user information; a processing unit that executes processing corresponding to the processing instruction information based on the usage information, using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; An information processing system having the above configuration.

[0105] The embodiments disclosed herein are illustrative in all respects and should not be considered as limiting. The scope of the present invention is indicated by the claims, not by the above meaning, and is intended to include all modifications within the meaning and scope of the claims. The present invention is not limited to the above-mentioned embodiments, and various modifications are possible within the scope of the claims, and embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0106] 10. Information Processing Systems 12 Server (information processing device) 22 Memory section 40 Information Processing Systems 50 Server (information processing device) 60 Storage section 220 Programs 222 trained models 226 LV1 Usage Information DB226 (Storage) 228 LV2 Usage Information DB228 (Storage) 600 Programs Pt Prompt (processing instruction information) Ot Processing result (processing execution result)

Claims

1. An information processing method executed by an information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a user information acquisition step of acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information, a direction of an output result based on the processing instruction information, and the user information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; An information processing method comprising:

2. The user information is associated with predetermined authority information that is set to at least one of a plurality of levels, the information source is stored in one of a plurality of storage units in which access permission is set for each level of the predetermined authority information, In the usage information acquisition step, the usage information is acquired from the information source in the storage unit that is accessible based on the predetermined authority information of the user information. The information processing method according to claim 1 .

3. a determination result acquisition step of acquiring a determination result as to whether or not the execution result of the processing in the processing step is satisfactory with respect to the processing instruction information based on a predetermined criterion; a good result storage step of storing the execution result of the process determined to be good in a storage unit in association with the processing instruction information; Including, When the correlation between the processing instruction information acquired in the processing instruction acquisition step and the processing instruction information stored in the good result storage step is equal to or greater than a predetermined standard, an execution result of the processing associated with the processing instruction information stored in the good result storage step is output. The information processing method according to claim 1 .

4. a separate processing step of executing a process based on the processing instruction information by a means separate from the process in the processing step, When the execution result of the processing executed in the separate processing step satisfies a predetermined condition, the execution result of the processing in the separate processing step is output instead of the execution result of the processing in the processing step. The information processing method according to claim 1 .

5. In the information processing device, a processing instruction acquisition step of acquiring processing instruction information instructing a processing; a user information acquisition step of acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition step of acquiring usage information required for processing from an information source based on the processing instruction information, a direction of an output result based on the processing instruction information, and the user information; a processing step of executing a process corresponding to the processing instruction information based on the usage information, using a trained model that trains training data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; A program for executing an information processing method including the steps of:

6. An information processing system executed by an information processing device, a processing instruction acquisition unit that acquires processing instruction information that instructs processing; a user information acquiring unit for acquiring user information relating to a user who inputs the processing instruction information; a usage information acquisition unit that acquires usage information required for processing from an information source based on the processing instruction information, a direction of an output result based on the processing instruction information, and the user information; a processing unit that executes processing corresponding to the processing instruction information based on the usage information, using a trained model that trains learning data including text data and outputs a processing result corresponding to the processing instruction information when the processing instruction information is input; An information processing system having the above configuration.

Images