Information processing device, information processing method and information processing program

Abstract

To appropriately classify post information.SOLUTION: An information processing device according to the present invention includes an extracting unit, a generating unit, and a providing unit. The extracting unit extracts, from a post information group posted by users, post information corresponding to a specified search query. The generating unit generates an instruction sentence that includes the post information extracted by the extracting unit and a letter string for sorting the post information. The providing unit provides a generative model that generates a reply sentence for an input sentence with the reply sentence obtained by inputting the instruction sentence generated by the generating unit.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

[0002] Conventionally, there is a technology that collects posted information on SNS (Social Network Service) and provides evaluation results for evaluation targets such as corporate brands and products. With such a technology, descriptions related to the evaluation target are extracted from the posted information, and evaluation is performed by linguistic and statistical analysis (for example, see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2015-135660 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional technology merely provides evaluation results regarding the evaluation target, and there is room for improvement in terms of appropriately classifying posted information.

[0005] The present invention has been made in consideration of the above, and has an object to provide an information processing device, an information processing method, and an information processing program that are capable of appropriately classifying posted information. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, an information processing device according to the present invention includes an extraction unit, a generation unit, and a provision unit. The extraction unit extracts posted information corresponding to a specified search query from a group of posted information posted by a user. The generation unit generates an instruction sentence including the posted information extracted by the extraction unit and a character string for classifying the posted information. The provision unit provides an answer sentence obtained by inputting the instruction sentence generated by the generation unit to a generation model that generates an answer sentence to an input sentence. Effect of the Invention

[0007] According to the present invention, appropriate content can be provided to a user. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. [Diagram 2] FIG. 2 is a diagram illustrating an example of an instruction sentence according to the embodiment. [Diagram 3] FIG. 3 is a diagram illustrating an example of a provision screen according to the embodiment. [Figure 4] FIG. 4 is a block diagram illustrating an example of the configuration of the information processing device according to the embodiment. [Diagram 5] FIG. 5 is a diagram illustrating an example of information stored in a posted information storage unit according to the embodiment. [Figure 6] FIG. 6 is a flowchart illustrating an example of a providing process according to the embodiment. [Figure 7] FIG. 7 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the information processing device, the information processing method, and the information processing program according to the present application will be described in detail with reference to the drawings. Note that the information processing device, the information processing method, and the information processing program according to the present application are not limited to the embodiments.

[0010] [Embodiment] [1. Information Processing] First, an example of information processing according to the embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of information processing according to the embodiment. Note that the information processing according to the embodiment is realized by an information processing device 1 shown in Fig. 1.

[0011] 1 is an information processing device that classifies posted information posted to a social network system (SNS) server 200. For example, the information processing device 1 classifies posted information related to a defect of a specified evaluation target from among a group of posted information posted to the SNS server 200, using a generative model.

[0012] The generative model is a model trained to generate answer sentences to questions, and is, for example, a Generative Pretrained Transformer (GPT) model. Note that the generative model may also be a natural language processing model such as Transformer.

[0013] The service provider terminal 100 is a terminal of a provider that provides a predetermined service. For example, the service provider terminal 100 requests the information processing device 1 to provide posted information related to the company's products and services, and provides (displays) the request result to the service provider. In this embodiment, the predetermined service is assumed to be a service related to electronic money payment.

[0014] The SNS server 200 shown in Fig. 1 is a server device that operates various SNSs (Social Networking Services). Various types of information are posted to the SNS by each user. Therefore, the posted information allows service providers to obtain feedback on their services and products.

[0015] On the other hand, since a huge amount of posted information is posted to the SNS server 200, it is not easy to classify the posted information desired by the service provider. Therefore, the information processing device 1 according to the embodiment classifies the posted information using a generative model. Note that, hereinafter, a case will be described in which posted information related to dissatisfaction with a service provided by a service provider (company) is classified from the posted information. Note that the posted information includes various information posted by general users, and includes, for example, posts to word-of-mouth sites, posts to news articles, etc.

[0016] 1, the information processing device 1 receives a request to categorize posted information from a service provider via the service provider terminal 100 (step S1). The categorization request includes a search query corresponding to a service or product for which the service provider requests feedback.

[0017] Next, the information processing device 1 extracts posted information corresponding to the search query specified by the service provider from a group of posted information posted to the SNS server 200 (step S2). From the group of posted information, the information processing device 1 extracts, for example, posted information that was posted during a period specified by the service provider and that is a hit for the search query.

[0018] Next, the information processing device 1 estimates the poster's emotion for each piece of extracted posted information (step S3). For example, the information processing device 1 estimates the poster's emotion from each piece of posted information using an estimation model trained to estimate the poster's emotion from the text of the posted information. For example, the estimation model is a model trained to classify the poster's emotion into about 30 emotions when posted information is input, and output a numerical value for each emotion.

[0019] Next, the information processing device 1 filters the posted information based on the poster's emotion (step S4). In this embodiment, in order to classify posted information related to dissatisfaction with the service, the information processing device 1 further extracts, from the extracted posted information, posted information whose poster's emotion is negative (for example, anger or sadness) by filtering. More specifically, the information processing device 1 extracts, by filtering, posted information whose poster's anger value exceeds a threshold value.

[0020] In this way, the information processing device 1 can reduce the information to be input to the generative model by selecting posted information whose poster's emotion is negative (for example, anger or sadness) from the posted information corresponding to the search query, thereby suppressing the processing load of the generative model. Note that the information processing device 1 may input all posted information to the generative model without filtering the posted information by emotion.

[0021] Next, the information processing device 1 generates an instruction sentence (step S5). The instruction sentence includes posted information filtered according to emotions and a character string indicating an instruction for classifying the posted information into a generative model. A specific example of the instruction sentence will be described later with reference to FIG. 2.

[0022] Then, the information processing device 1 inputs the generated instruction sentence into the generation model, and provides information on the answer sentence output from the generation model to the service provider terminal 100 (step S6). Note that a display screen provided to the service provider will be described later with reference to FIG.

[0023] In this way, the information processing device 1 according to the embodiment extracts posted information corresponding to the search query from a group of posted information, and classifies the posted information using a generative model. Therefore, the information processing device 1 according to the embodiment can appropriately classify the posted information.

[0024] Next, an example of an instruction sentence according to the embodiment will be described with reference to Fig. 2. Fig. 2 is a diagram showing an example of an instruction sentence according to the embodiment. As shown in Fig. 2, an instruction sentence A includes a character string Ta and posted information Tw.

[0025] The character string Ta is a character string related to an instruction for classifying the posted information. In Fig. 2, the character string Ta is "Please select one of the lower categories to classify the following text. Categories: Service bugs, complaints about the service, questions, troubles between users". The information processing device 1 may be configured to receive a designation of the character string Ta from the service provider.

[0026] The posted information Tw included in the instruction sentence A is the posted information to be classified, and is posted information corresponding to the search query, which is posted information whose poster's sentiment is negative. Note that while the posted information Tw is abstractly shown in FIG. 2 as "posted information #1," "posted information #2," and "posted information #3," it is assumed that the posted information Tw includes the posted information actually posted by each poster.

[0027] Next, an example of a provision screen provided to a service provider will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of a provision screen according to an embodiment. For example, as shown in Fig. 3, the provision screen G has items such as "category", "post", "emotion", and "posting date and time".

[0028] "Category" indicates the category of the corresponding posted information. In FIG. 3, examples of categories include "dissatisfaction with the service" and "service bugs." It should be noted that the categories are not limited to the above examples, and may include other items such as troubles between users. "Post" indicates the corresponding posted information. "Emotion" indicates the emotion of the poster who posted the corresponding posted information. It should be noted that in FIG. 3, "emotion" is shown as a numerical value for the emotion of anger.

[0029] "Posting date and time" indicates the date and time when the corresponding post information was posted. In this manner, the provision screen G provides information regarding each category, post, emotion, and posting date and time to the service provider.

[0030] This allows the service provider to check posted information regarding dissatisfaction with the service provided by the service provider by referring to the provision screen G. Note that the provision screen G shown in Fig. 3 is only an example, and the content and layout of the information to be provided may be changed as desired.

[0031] [2. Information Processing Device] Next, a configuration example of the information processing device 1 according to the embodiment will be described with reference to Fig. 4. Fig. 4 is a block diagram showing a configuration example of the information processing device 1 according to the embodiment. As shown in Fig. 4, the information processing device 1 includes a communication unit 2, a storage unit 3, and a control unit 4. Note that the information processing device 1 may also include an input unit (e.g., a keyboard, a mouse, etc.) that receives various operations from an administrator who uses the information processing device 1, and a display unit (e.g., a liquid crystal display, etc.) that displays various information.

[0032] The communication unit 2 is realized by, for example, a network interface card (NIC) etc. The communication unit 2 is connected to a communication network such as 4G (4th Generation) or 5G (5th Generation) by wire or wirelessly, and transmits and receives information to and from the service provider terminal 100, the SNS server 200, etc. via the communication network.

[0033] The storage unit 3 is realized by, for example, a semiconductor memory element such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 3 includes a posted information storage unit 31, an estimation model storage unit 32, and a generation model storage unit 33.

[0034] The posted information storage unit 31 stores information related to posted information posted to the SNS server 200. Fig. 5 is a diagram showing an example of information stored in the posted information storage unit 31 according to the embodiment. As shown in Fig. 5, the posted information storage unit 31 stores information items such as "posting ID", "poster information", "posting content", and "posting date and time" in association with each other.

[0035] The "posting ID" is an identifier for identifying each piece of posted information. The "poster information" is information about the poster who posted using the corresponding posting ID, such as account information of the poster in the SNS server 200.

[0036] The "posted content" is the posted content related to the posted information identified by the corresponding post ID, and includes text and images posted by the poster. Note that the "posted content" may also include content related to replies to the posted information.

[0037] The "posting date and time" is the date and time when the post information identified by the corresponding post ID was posted. In addition, when the location where the post information was posted can be specified, the post information storage unit 31 may store information related to the location in association with the post information.

[0038] Returning to the description of FIG. 4, the estimation model storage unit 32 will be described. The estimation model storage unit 32 stores an estimation model. The estimation model is a model trained to estimate the poster's emotion from the posted information. For example, in a learning stage, the estimation model learns using correct answer data that links the posted information with the poster's emotion.

[0039] The generative model storage unit 33 stores the generative model. The generative model is, for example, a GPT model, which is a model that generates an answer sentence to a question. In this embodiment, when an instruction sentence is input, the generative model classifies posted information included in the instruction sentence. Note that the information processing device 1 may use models stored in an external server for the estimation model and the generative model.

[0040] Next, the control unit 4 will be described. The control unit 4 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs (corresponding to an example of an information processing program) stored in a storage device inside the information processing device 1 using a RAM as a working area. The control unit 4 is also, for example, a controller, and is realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0041] As shown in Fig. 4, the control unit 4 includes a reception unit 41, an extraction unit 42, an estimation unit 43, a generation unit 44, and a provision unit 45, and realizes or executes the functions and actions of information processing described below. The internal configuration of the control unit 4 is not limited to the configuration shown in Fig. 4, and may be other configurations as long as they perform the information processing described below. Furthermore, the connection relationship between the processing units in the control unit 4 is not limited to the connection relationship shown in Fig. 4, and may be other connection relationships.

[0042] The reception unit 41 receives, for example, a request to categorize posted information from a service provider via the service provider terminal 100. For example, the categorization request includes information on a search query (evaluation target), categorization content (e.g., category to categorize), poster's emotion to be categorized, posting date and time, etc. Then, the reception unit 41 passes the information on the categorization request received from the service provider to the poster.

[0043] The extraction unit 42 extracts posted information corresponding to the specified search query from a group of posted information posted by a user (contributor). For example, the extraction unit 42 extracts posted information corresponding to the search query included in the classification request received from the reception unit 41 from the posted information posted to the SNS server 200.

[0044] At this time, if the service provider specifies the posting date and time, the extraction unit 42 extracts the posted information for the specified period. Then, the extraction unit 42 stores the posted information extracted based on the search query in the posted information storage unit 31.

[0045] The estimation unit 43 estimates the emotion of the poster who posted the posted information from the posted information. For example, the estimation unit 43 estimates the emotion of the poster of the posted information extracted by the extraction unit 42 by using the estimation model stored in the estimation model storage unit 32.

[0046] For example, the estimation unit 43 inputs each piece of posted information extracted by the extraction unit 42 to an estimation model, and estimates the emotion of each poster based on the output result of the estimation model. Then, the estimation unit 43 filters each piece of posted information according to the emotion.

[0047] For example, the estimation unit 43 extracts, by filtering, posted information with a negative emotion from the posted information extracted by the extraction unit 42. Then, the estimation unit 43 passes information on the posted information extracted by filtering to the generation unit 44. Note that the estimation unit 43 can arbitrarily change the emotion when filtering, for example, by extracting posted information with an emotion designated by a service provider.

[0048] The generation unit 44 generates an instruction sentence including the posted information extracted by the extraction unit 42 and a character string for classifying the posted information. The generation unit 44 generates the instruction sentence by classifying, among the posted information extracted by the extraction unit 42, posted information in which the poster's emotion estimated by the estimation unit belongs to a predetermined category.

[0049] In the present embodiment, the generation unit 44 generates an instruction sentence that classifies the posted information related to the defect of the evaluation target when the poster's emotion is negative. Specifically, the generation unit 44 generates an instruction sentence A as shown in FIG.

[0050] The generating unit 44 may generate the instruction sentence A including different character strings depending on the emotion. For example, when a service provider requests that the posted information be classified into those that users feel favorably about the service of the service provider, the generating unit 44 generates the instruction sentence A including character strings for such classification.

[0051] In this case, the generation unit 44 may generate instruction sentence A including a character string selected by the service provider from a plurality of character strings registered in advance, or may generate instruction sentence A including a character string input by the service provider, for example.

[0052] Then, the generation unit 44 passes the generated instruction sentence A to the provision unit 45. Note that the generation unit 44 may classify the posted information according to the poster, for example, and generate the instruction sentence A for each of the classified posted information.

[0053] For example, the generation unit 44 may estimate the attributes of each poster from the history of the posted information of each poster, classify the posted information for each attribute of the poster, and generate each instruction sentence A. The attributes of the poster include, for example, occupation, age, annual income, family structure, place of residence, etc. For example, the generation unit 44 converts each piece of posted information posted by each poster into a feature vector, and estimates the attribute of the poster according to the feature vector.

[0054] In this way, the generation unit 44 can provide more detailed feedback to the service provider by generating the instruction sentence A for each attribute of the contributor.

[0055] The providing unit 45 provides an answer sentence obtained by inputting an instruction document generated by the generating unit 44 to a generative model that generates an answer sentence to a question. For example, the providing unit 45 inputs instruction sentence A generated by a generative model to a generative model stored in the generative model storage unit 33, and receives an answer to instruction sentence A from the generative model.

[0056] Then, the providing unit 45 generates content related to the answer received from the generative model, and provides the content to the service provider terminal 100. For example, the content related to the answer includes information on items such as "category", "post", "emotion", and "posting date and time" as shown in FIG.

[0057] This allows the service provider to easily obtain posted information such as each user's dissatisfaction with the service of the service provider. More specifically, in the conventional method, the posted information is subjected to morphological analysis, syntactic analysis, etc., and character strings defined in advance as evaluation expressions are detected from the analysis results, thereby classifying the desired posted information. Therefore, there is a problem that the accuracy of classifying the posted information is low. In contrast, the method according to the present invention applies natural language processing of a generative model, and thereby makes it possible to obtain an output with an accuracy closer to the accuracy of classification performed by a human being after understanding the content of the posted information.

[0058] [3. Processing flow] Next, a process procedure executed by the information processing device 1 according to the embodiment will be described with reference to Fig. 6. Fig. 6 is a flowchart showing an example of a providing process according to the embodiment. Note that the process procedure shown below is repeatedly executed by the information processing device 1 upon receiving a classification request from a service provider.

[0059] 6, the information processing device 1 first extracts posted information corresponding to the search query from a group of posted information (step S101). Next, the information processing device 1 estimates the poster's emotion for each piece of extracted posted information (step S102).

[0060] Next, the information processing device 1 filters the posted information based on the estimated emotion (step S103). Next, the information processing device 1 generates an instruction sentence based on the filtered posted information (step S104), and inputs the generated instruction sentence into a generative model (step S105).

[0061] Then, the information processing device 1 provides the answer sentence output from the generative model to the service provider (step S106), and ends the process.

[0062] 4. Modifications In the above-described embodiment, a case has been described in which the information processing device 1 classifies posted information related to dissatisfaction with an evaluation target (search query) specified by a service provider using a generative model, but this is not limited to this.

[0063] For example, the evaluation target may be any object such as a store, a region, or a person, and further, the information processing device 1 may generate an instruction sentence to classify favorable posted information about the evaluation target or neutral posted information.

[0064] 5. Effects The information processing device 1 according to the embodiment includes an extraction unit 42 that extracts posted information corresponding to a specified search query from a group of posted information posted by a user, a generation unit 44 that generates an instruction sentence including the posted information extracted by the extraction unit 42 and a character string for classifying the posted information, and a provision unit 45 that provides the answer sentence obtained by inputting the instruction document generated by the generation unit 44 to a generation model that generates an answer sentence to a question.

[0065] Furthermore, the generation unit 44 generates an instruction sentence including a character string for classifying the posted information extracted by the extraction unit 42 into a plurality of categories. Furthermore, the information processing device 1 includes an estimation unit 43 that estimates the emotion of a poster who posted the posted information from the posted information, and the generation unit 44 generates an instruction sentence for classifying posted information, among the posted information extracted by the extraction unit 42, of which the emotion of the poster estimated by the estimation unit 43 belongs to a predetermined category.

[0066] The generation unit 44 generates instruction sentences including different character strings depending on the emotion. The extraction unit 42 extracts posted information corresponding to a search query indicating an evaluation target, and the generation unit 44 generates instruction sentences including character strings that classify the posted information into information related to a defect of the evaluation target when the poster's emotion is negative. The generation model is a GPT (Generative Pretrained Transformer) model.

[0067] By performing any one or a combination of the above-described processes, the information processing device according to the present application can appropriately classify posted information.

[0068] [6. Hardware Configuration] The information processing device 1 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in Fig. 7. Fig. 7 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device according to the embodiment. The computer 1000 has a CPU 1100, a RAM 1200, a ROM 1300, a HDD 1400, a communication interface (I / F) 1500, an input / output interface (I / F) 1600, and a media interface (I / F) 1700.

[0069] The CPU 1100 operates and controls each unit based on a program stored in the ROM 1300 or the HDD 1400. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, programs that depend on the hardware of the computer 1000, and the like.

[0070] The HDD 1400 stores programs executed by the CPU 1100, data used by such programs, etc. The communication interface 1500 receives data from other devices via a network (communication network) N and sends the data to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the network N.

[0071] The CPU 1100 controls output devices such as a display and a printer, and input devices such as a keyboard and a mouse (in FIG. 7, the output devices and the input devices are collectively referred to as "input / output devices") via an input / output interface 1600. The CPU 1100 acquires data from the input devices via the input / output interface 1600. The CPU 1100 also outputs generated data to the output devices via the input / output interface 1600.

[0072] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0073] For example, when the computer 1000 functions as an information processing device according to the embodiment, the CPU 1100 of the computer 1000 executes programs loaded onto the RAM 1200 to realize the functions of the control unit 4. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, the CPU 1100 may obtain these programs from another device via the network N.

[0074] [7. Other] Although the embodiments of the present application have been described above, the present invention is not limited to the contents of these embodiments. The above-described components include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the so-called equivalent range. Furthermore, the above-described components can be appropriately combined. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the spirit of the above-described embodiments.

[0075] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by a known method. In addition, the information including the processing procedures, specific names, various data and parameters shown in the above documents and drawings can be changed arbitrarily unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0076] In addition, each component of each device shown in the figure is a functional concept, and does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads, usage conditions, etc.

[0077] For example, the above-mentioned information processing device may be realized using multiple server computers, and depending on the functions, the configuration can be flexibly changed, such as by calling an external platform using an API (Application Programming Interface) or network computing.

[0078] Furthermore, the above-described embodiments and modifications can be appropriately combined as long as the processing contents are not contradictory.

[0079] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit", etc. For example, an acquisition section can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0080] 1. Information processing device 2. Communications Department 3 Storage section 4. Control Unit 31 Posted information storage unit 32 Estimation model memory unit 33 Generative model storage unit 41 Reception 42 Extraction part 43 Estimation part 44 Generation part 45 Providing Department 100 Service provider terminal 200 SNS Server

Claims

1. an extraction unit that extracts posted information corresponding to a specified search query from a group of posted information posted by a user; a generation unit that generates an instruction sentence including the posted information extracted by the extraction unit and a character string for classifying the posted information; a providing unit that provides an answer sentence obtained by inputting the instruction sentence generated by the generating unit to a generation model that generates an answer sentence for an input sentence; An information processing device comprising:

2. The generation unit is generating the instruction sentence including the character string that causes the posted information extracted by the extraction unit to be classified into a plurality of categories; 2. The information processing device according to claim 1,

3. An estimation unit that estimates the emotion of a poster who posted the posted information from the posted information. Equipped with The generation unit is generating the instruction sentence by classifying, among the posted information extracted by the extraction unit, posted information in which the poster's emotion estimated by the estimation unit belongs to a predetermined category; 2. The information processing device according to claim 1,

4. The generation unit is generating instruction sentences including the character strings that differ according to the emotion; 4. The information processing apparatus according to claim 3,

5. The extraction unit is extracting the posted information corresponding to the search query indicating an evaluation target; The generation unit is generating the instruction sentence including the character string indicating that the poster's sentiment is negative and that the posted information is classified into the posted information related to the defect to be evaluated; 4. The information processing apparatus according to claim 3,

6. The generative model is It is a GPT (Generative Pretrained Transformer) model.

2. The information processing device according to claim 1,

7. 1. A computer-implemented information processing method, comprising: an extraction step of extracting posted information corresponding to the specified search query from a group of posted information posted by a user; a generating step of generating an instruction sentence including the posted information extracted by the extracting step and a character string for classifying the posted information; a providing step of providing an answer sentence obtained by inputting the instruction sentence generated by the generating step to a generation model that generates an answer sentence for the input sentence; 13. An information processing method comprising:

8. An extraction step of extracting posted information corresponding to a specified search query from a group of posted information posted by a user; a generating step of generating an instruction sentence including the posted information extracted by the extracting step and a character string for classifying the posted information; a provision step of providing an answer sentence obtained by inputting the instruction sentence generated by the generation step to a generation model that generates an answer sentence for the input sentence; An information processing program characterized by causing a computer to execute the above.

Images