Information processing device, information processing method, and information processing program

Abstract

User comments are presented in an appropriate writing style for each category. [Solution] The information processing device according to the present invention is characterized by having an estimation unit that estimates the category of comments posted by users to news articles (for example, an opinion category that shows the user's opinion, or a fact category that shows facts about the events shown in the news article), and a generation unit that generates text information that aggregates comments, which is text information that shows the comments in a style corresponding to the category estimated by the estimation unit.

Description

Technical Field

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

Background Art

[0002] Conventionally, techniques related to information distribution via the Internet are known. As an example of such a technique, there is known a technique of receiving submissions of articles from a plurality of users via a network, editing the articles in the form of an electronic newspaper, and publishing them.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, with the above-described technique, it cannot be said that comments from users can be provided in an appropriate style according to the category.

[0005] For example, with the above-described technique, it only publishes submissions from users, and it cannot be said that comments from users can be provided in an appropriate style according to the category.

[0006] The present application has been made in view of the above, and an object thereof is to provide comments from users in an appropriate style according to the category.

Means for Solving the Problems

[0007] The information processing device according to the present invention is characterized by having an estimation unit that estimates the category of comments posted by users to news articles, and a generation unit that generates text information that aggregates the comments, which is text information that represents the comments in a style corresponding to the category estimated by the estimation unit. [Effects of the Invention]

[0008] According to one embodiment of the system, it is possible to provide user comments in an appropriate writing style according to the category. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 shows an example of information processing according to the embodiment. [Figure 2] Figure 2 shows an example of the configuration of the information processing device 10 according to the embodiment. [Figure 3] Figure 3 shows an example of a news article database 31. [Figure 4] Figure 4 shows an example of the comment database 32. [Figure 5] Figure 5 is a flowchart showing an example of the information processing procedure according to the embodiment. [Figure 6] Figure 6 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10. [Modes for carrying out the invention]

[0010] The following describes in detail, with reference to the drawings, the embodiments for implementing the information processing device, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments"). Note that these embodiments do not limit the information processing device, information processing method, and information processing program according to the present application. Furthermore, the same parts are denoted by the same reference numerals in each of the following embodiments, and redundant descriptions are omitted.

[0011] [1. Embodiments] The information processing realized by the information processing device of this embodiment will be explained using Figure 1. Figure 1 is a diagram showing an example of information processing according to the embodiment. In Figure 1, the information processing according to the embodiment is realized by the information processing device 10, which is an example of the information processing device according to the present application.

[0012] As shown in Figure 1, the information processing system 1 according to this embodiment includes an information processing device 10 and a user terminal 100. The information processing device 10 and the user terminal 100 are connected to each other via a network N (see, for example, Figure 2) by wire or wireless means so that they can communicate with each other. The network N is, for example, a WAN (Wide Area Network) such as the Internet. Note that the information processing system 1 shown in Figure 1 may include multiple information processing devices 10 and multiple user terminals 100.

[0013] The information processing device 10 shown in Figure 1 is an information processing device that realizes information processing according to the embodiment, and is realized by, for example, a server device or a cloud system. In the example in Figure 1, the information processing device 10 is an information processing device that provides a news distribution service that distributes content showing news articles collected from various mass media such as newspaper companies, publishing companies, and broadcasting stations (television stations, radio stations).

[0014] The information processing device 10 may also function as a web server providing a website related to the news service. Furthermore, the information processing device 10 may be a device that distributes information to the user terminal 100 for display on an application related to the news service (hereinafter sometimes referred to as "news app") installed on the user terminal 100. The information processing device 10 may also be a server that distributes the news app data itself. Furthermore, the information processing device 10 may function as a distribution device that distributes control information to the user terminal 100. Here, the control information is written in, for example, a scripting language such as JavaScript®, a stylesheet language such as CSS (Cascading Style Sheets), or a programming language such as Java®, Kotlin, Swift, or Objective-C. The news app itself distributed from the information processing device 10 may also be considered as control information.

[0015] The user terminal 100 shown in Figure 1 is an information processing device used by a user. The user terminal 100 can be implemented as, for example, a smartphone, a tablet, a notebook PC (Personal Computer), a desktop PC, a mobile phone, or a PDA (Personal Digital Assistant). In the example shown in Figure 1, the user terminal 100 is a smartphone used by the user.

[0016] Furthermore, the user terminal 100 displays the information provided by the information processing device 10 using a web browser or application. When the user terminal 100 receives control information from the information processing device 10 or the like to implement the information display process, it implements the display process according to the control information.

[0017] Next, the information processing performed by the information processing apparatus 10 will be described using FIG. 1. In the following description, the user terminals 100-1 to 100-N (N is an arbitrary natural number) will be described according to the users who use the user terminal 100. For example, the user terminal 100-1 is the user terminal 100 used by the user (user U1) identified by the user ID "UID#1". Also, in the following, when the user terminals 100-1 to 100-N are described without particular distinction, they will be referred to as the user terminal 100. Also, in the following description, the user terminal 100 may be regarded as the same as the user. That is, in the following, the user can also be read as the user terminal 100.

[0018] Also, in the following description, it is assumed that a news app is installed on the user terminal 100.

[0019] First, the information processing apparatus 10 receives, from the user terminal 100, the posting of a user's comment (for example, text information) on a news article provided via the news app (step S1). In the example of FIG. 1, it is assumed that the information processing apparatus 10 has received comments C1 "AA thinks BB", C2 "CC should have been DD",... as the user's comments on the news article #1 regarding event #1 (for example, an incident, an accident, a disaster, etc.).

[0020] Subsequently, the information processing apparatus 10 estimates the categories of comments C1, C2,... (step S2). For example, the information processing apparatus 10 estimates the category of each of the comments C1, C2,... based on the context indicated by each comment.

[0021] To give a specific example, the information processing apparatus 10 inputs comment C1 and an instruction sentence that instructs the model #1, which has been learned to generate an answer to the input question, to output the category of comment C1 based on the context indicated by comment C1, thereby estimating the category of comment C1. Also, the information processing apparatus 10 estimates the category of each comment on the news article #1 using the same method.

[0022] To give a more specific example, the information processing device 10 estimates whether each category of comments C1, C2, ... belongs to an opinion category indicating the user's opinion, or a fact category indicating a fact about event #1 as described in news article #1. For example, if the comment ends with a conjecture such as "I think..." or "It might be...", the information processing device 10 estimates that the category of the comment is an opinion category indicating the user's opinion (or impression). Also, if the comment ends with a definitive statement such as "It is..." or "It is likely...", the information processing device 10 estimates that the category of the comment is a fact category indicating a fact about event #1 as described in news article #1.

[0023] The information processing device 10 may also estimate each category of comments C1, C2, ... based on the corresponding news article that corresponds to news article #1. Here, the corresponding news article is a predetermined news article relating to event #1, and may be, for example, a news article that shows primary information relating to event #1 (for example, a news article showing a press release), or a news article that has been set in advance by the administrator of news service #1, etc.

[0024] For example, the information processing device 10 estimates the category of comment C1 by inputting comment C1, corresponding news article information (e.g., text information describing the content of the corresponding news article) related to corresponding news article #1, and an instruction sentence instructing the model #2, which has been trained to generate answers to input questions, to output the category of comment C1 based on the corresponding news article information. The information processing device 10 also estimates the category of each comment for news article #1 using a similar method.

[0025] For example, if the content of a comment is included in the corresponding news article, the information processing device 10 assumes that the category of the comment is a factual category. Furthermore, if the content of a comment is not included in the corresponding news article, the information processing device 10 assumes that the category of the comment is an opinion category.

[0026] Next, the information processing device 10 generates text information that represents comments C1, C2, ... in a style corresponding to the categories of comments C1, C2, ... and aggregates comments C1, C2, ... (step S3). For example, the information processing device 10 generates text information T1 by inputting a model #3, which has been trained to generate answers to an input question, comments C1, C2, ..., information representing each category of comments C1, C2, ..., and an instruction sentence instructing the model to output text information that aggregates comments C1, C2, ... in a style corresponding to the categories.

[0027] For example, if comment C1 is estimated to be an opinion category, the information processing device 10 generates text information T1 that indicates comment C1 in a style that makes it clear that the content of comment C1 is not factual but rather the user's opinion (or impression). For example, the information processing device 10 generates text information T1 that includes the string "There is an opinion that AA is BB" corresponding to comment C1.

[0028] Furthermore, if comment C2 is presumed to be in the fact category, the information processing device 10 generates text information T1 that indicates comment C2 in a style that makes it clear that the content of comment C2 is factual. For example, the information processing device 10 generates text information T1 that includes the string "CC is reported to be DD" corresponding to comment C2.

[0029] In other words, the information processing device 10 modifies the endings of the comments and generates aggregated text information T1 so that the category of the comments can be identified.

[0030] Models #1 to #3 described above are trained to output answers corresponding to input questions, and are large-scale language models (LLMs) that perform natural language processing, such as GPT (Generative Pre-trained Transformer) and Transformer. Models #1 to #3 reside within the information processing device 10 and were created independently by the business operator managing the information processing device 10. It is desirable that the input information be kept confidential by training the models so that it is not used as a new answer.

[0031] Next, the information processing device 10 provides text information T1 to the user terminal 100 via the news application (step S4). For example, the information processing device 10 provides text information T1 along with news article #1 as a summary of user comments on news article #1. The information processing device 10 also provides text information T1 as a news article related to event #1 (for example, a news article showing user reactions to event #1).

[0032] As described above, the information processing device 10 according to the embodiment estimates the category of a user's comment on a news article and generates text information representing each comment in a style corresponding to the estimated category. This allows the information processing device 10 to provide user comments in an appropriate style according to their category.

[0033] Furthermore, comments posted by users on news articles can sometimes contain a mix of facts and the user's opinions (or impressions). Therefore, in the past, in order for users to understand whether a comment represents a fact or an opinion, it was necessary to categorize the comments and present them to the user, or to attach icons to the comments to indicate the category.

[0034] In response to this, the information processing device 10 according to the embodiment estimates the category of the comment and generates text information with the style of the comment modified so that the category can be understood. As a result, the information processing device 10 according to the embodiment makes it possible for the user to naturally understand whether the comments constituting the text information are facts or the user's opinion simply by viewing the text information.

[0035] [2. Other processing examples] The process described above is merely one example, and the information processing device 10 may perform various processes using various types of information. Examples of this are listed below.

[0036] [2-1. About Categories] In the example in Figure 1, the category of comments estimated by the information processing device 10 is not limited to those that express facts or opinions. For example, the information processing device 10 may estimate whether a comment belongs to a positive category, indicating that it is positive about the events described in the news article, or a negative category, indicating that it is negative about the events described in the news article. The information processing device 10 may then generate text information T1 in which the ending of a comment estimated to belong to the positive category has been changed to "~there are also voices of support such as..." The information processing device 10 may also generate text information T1 in which the ending of a comment estimated to belong to the negative category has been changed to "~there are also harsh voices such as..."

[0037] [3. Configuration of the Information Processing Device] Next, the configuration of the information processing device 10 will be described using Figure 2. Figure 2 is a diagram showing an example of the configuration of the information processing device 10 according to the embodiment. As shown in Figure 2, the information processing device 10 has a communication unit 20, a storage unit 30, and a control unit 40.

[0038] (Regarding Communications Section 20) The communication unit 20 is implemented, for example, by a NIC (Network Interface Card). The communication unit 20 is connected to the network N by wire or wireless connection and transmits and receives information with the user terminal 100, etc.

[0039] (Regarding memory unit 30) The storage unit 30 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as hard disks and optical discs. As shown in Figure 2, the storage unit 30 has a news article database 31, a comment database 32, and a model database 33.

[0040] (Regarding News Article Database 31) The news article database 31 stores various types of information related to news articles. Here, an example of the information stored in the news article database 31 is explained using Figure 3. Figure 3 is a diagram illustrating an example of the news article database 31. In the example in Figure 3, the news article database 31 has items such as "news article ID," "event information," "news article text information," and "comment information."

[0041] "News Article ID" indicates identification information used to identify a news article. "Event Information" indicates information about the event described in the news article, such as identification information to identify the event and text information describing the content of the event. "News Article Text Information" indicates text information of the content of the news article (such as the title and body). "Comment Information" indicates information about comments posted to the news article, such as identification information to identify the comment and text information describing the comment.

[0042] In other words, Figure 3 shows an example where the event information of a news article identified by the news article ID "NID#1" is "Event Information#1", the news article text information is "News Article Text Information#1", and the comment information is "Comment Information#1".

[0043] (Regarding comment database 32) The comment database 32 stores various information about user comments posted on news articles. Here, an example of the information stored in the comment database 32 is illustrated using Figure 4. Figure 4 is a diagram illustrating an example of the comment database 32. In the example in Figure 4, the comment database 32 has items such as "Comment ID," "News Article ID," "Comment Text Information," and "Category Information."

[0044] "Comment ID" indicates identification information used to identify the comment. "News Article ID" indicates identification information used to identify the news article to which the comment was posted. "Comment Text Information" indicates the text information describing the comment. "Category Information" indicates the category of the comment.

[0045] In other words, Figure 4 shows an example where a comment identified by comment ID "CID#1" is posted to a news article identified by news article ID "NID#1", and the comment text information of that comment is "comment text information#1" and the category information is "category information#1".

[0046] (Regarding Model Database 33) The model database 33 stores models that have been trained to generate answers to input questions. For example, the model database 33 stores models #1 through #3.

[0047] (Regarding the control unit 40) The control unit 40 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs stored in the memory device inside the information processing device 10 using RAM as a working area. Alternatively, the control unit 40 is a controller, and is realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). As shown in Figure 2, the control unit 40 according to this embodiment has an estimation unit 41, a generation unit 42, and a providing unit 43, and realizes or executes the information processing functions and operations described below.

[0048] (Regarding the estimation section 41) The estimation unit 41 estimates the category of comments posted by users to news articles. For example, in the example in Figure 1, the estimation unit 41 refers to the memory unit 30 (for example, the news article database 31 or the comment database 32) and estimates the categories of user comments C1, C2, ... for news article #1.

[0049] Furthermore, the estimation unit 41 may estimate the category of a comment based on the context in which the comment is expressed. For example, in the example in Figure 1, the estimation unit 41 estimates the category of each of the comments C1, C2, ... based on the context in which each comment is expressed.

[0050] Furthermore, the estimation unit 41 may estimate the category of a comment by inputting a comment and an instruction sentence that instructs the model, which has been trained to generate answers to input questions, to output the category of the comment based on the context in which the comment indicates. For example, in the example in Figure 1, the estimation unit 41 refers to the memory unit 30 (for example, a news article database 31, a comment database 32, a model database 33, etc.) and estimates the category of comment C1 by inputting comment C1 and an instruction sentence that instructs the model #1, which has been trained to generate answers to input questions, to output the category of comment C1 based on the context in which comment C1 indicates.

[0051] Furthermore, the estimation unit 41 may estimate the category of a comment based on information about the corresponding news article that corresponds to the news article. For example, in the example in Figure 1, the estimation unit 41 estimates the category of each of the comments C1, C2, ... based on the corresponding news article that corresponds to news article #1.

[0052] Furthermore, the estimation unit 41 may estimate the category of a comment by inputting a comment, information about the corresponding news article, and an instruction sentence instructing the model, which has been trained to generate an answer to an input question, to output the category of the comment based on the information about the corresponding news article. For example, in the example in Figure 1, the estimation unit 41 estimates the category of comment C1 by inputting comment C1, information about the corresponding news article #1, and an instruction sentence instructing the model #2, which has been trained to generate an answer to an input question, to output the category of comment C1 based on the information about the corresponding news article.

[0053] Furthermore, the estimation unit 41 may estimate whether a comment belongs to an opinion category indicating the user's opinion, or to a fact category indicating facts about the event described in the news article. For example, in the example in Figure 1, the estimation unit 41 estimates whether each category of comments C1, C2, ... belongs to an opinion category indicating the user's opinion, or to a fact category indicating facts about event #1 described in news article #1.

[0054] (Regarding the generation unit 42) The generation unit 42 generates text information that aggregates comments, which is text information that shows comments in a style corresponding to the category estimated by the estimation unit 41. For example, in the example in Figure 1, the estimation unit 41 refers to the storage unit 30 (for example, the comment database 32) and generates text information that aggregates comments C1, C2, ..., which is text information that shows comments C1, C2, ..., in a style corresponding to the category of comments C1, C2, ...

[0055] Furthermore, the generation unit 42 may generate text information by inputting information about the comments, the categories of the comments, and an instruction sentence instructing the model, which has been trained to generate answers to the input questions, to output text information that represents the comments in a style corresponding to the category. For example, in the example in Figure 1, the generation unit 42 refers to the storage unit 30 (for example, a comment database 32 or a model database 33, etc.) and generates text information T1 by inputting comments C1, C2, ..., information indicating the category of each of the comments C1, C2, ..., and an instruction sentence instructing the model #3, which has been trained to generate answers to the input questions, to output text information that aggregates the comments C1, C2, ... in a style corresponding to the category.

[0056] (Regarding Section 43) The provisioning unit 43 provides text information to the user as a news article relating to the event indicated in the news article. For example, in the example in Figure 1, the provisioning unit 43 provides text information T1 as a news article relating to event #1.

[0057] [4. Information Processing Flow] The information processing procedure of the information processing device 10 according to the embodiment will be explained using Figure 5. Figure 5 is a flowchart showing an example of the information processing procedure according to the embodiment.

[0058] As shown in Figure 5, the information processing device 10 determines whether it has received a comment for the news article (step S101). If no comment has been received (step S101; No), the information processing device 10 waits until a comment is received.

[0059] On the other hand, if a comment is accepted (step S101; Yes), the information processing device 10 estimates the category of the submitted comment (step S102). Subsequently, the information processing device 10 generates text information that aggregates the comments, which is text information that indicates the comments in a style corresponding to the category (step S103), and then terminates the process.

[0060] [5. Variations] The above-described embodiment is merely an example, and various modifications and applications are possible.

[0061] [5-1. Regarding the processing method] Of the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, and conversely, all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above text and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0062] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0063] Furthermore, the embodiments described above can be combined as appropriate, provided that the processing content is not contradictory.

[0064] [6. Effects] As described above, the information processing device 10 according to the embodiment includes an estimation unit 41, a generation unit 42, and a provision unit 43. The estimation unit 41 estimates the category of comments posted by users to news articles. The estimation unit 41 also estimates whether the comments belong to an opinion category indicating the user's opinion or a fact category indicating facts about the events described in the news article. The generation unit 42 generates text information that aggregates the comments, which is text information that shows the comments in a style corresponding to the category estimated by the estimation unit 41. The generation unit 42 also generates text information by inputting information about the comments, the category of the comments, and an instruction sentence that instructs a model trained to generate answers to input questions to output text information that shows the comments in a style corresponding to the category. The provision unit 43 provides the text information to the user as a news article about the events described in the news article.

[0065] As a result, the information processing device 10 according to this embodiment can estimate the category of a user's comment on a news article and generate text information that represents each comment in a style corresponding to the estimated category, thus enabling the provision of user comments in an appropriate style according to the category.

[0066] Furthermore, in the information processing device 10 according to the embodiment, for example, the estimation unit 41 estimates the category of a comment based on the context indicated by the comment. The estimation unit 41 also estimates the category of a comment by inputting the comment and an instruction sentence that instructs a model trained to generate answers to input questions to output the category of the comment based on the context indicated by the comment. Furthermore, the estimation unit 41 also estimates the category of a comment based on information about a corresponding news article that corresponds to a news article. Furthermore, the estimation unit 41 also estimates the category of a comment by inputting the comment and an instruction sentence that instructs a model trained to generate answers to input questions to output the category of the comment based on information about a corresponding news article.

[0067] As a result, the information processing device 10 according to this embodiment can accurately estimate the category of a comment based on the context of the comment and the results of comparing the comment with a news article, and can provide the comment in an appropriate style according to its category.

[0068] [7. Hardware Configuration] Furthermore, the information processing device 10 according to each embodiment described above can be implemented by a computer 1000 having a configuration such as that shown in Figure 6. The following explanation will use the information processing device 10 as an example. Figure 6 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10. The computer 1000 has a CPU 1100, ROM 1200, RAM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0069] The CPU 1100 operates based on programs stored in the ROM 1200 or HDD 1400, and controls various parts. The ROM 1200 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0070] The HDD 1400 stores programs executed by the CPU 1100, as well as data used by such programs. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to network N in this embodiment) and sends it to the CPU 1100, and also transmits data generated by the CPU 1100 to other devices via the communication network 500.

[0071] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the data it generates to 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 1300. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1300 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 PD (Phase Change Rewritable Disk), 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 computer 1000 functions as information processing device 10, the CPU 1100 of computer 1000 realizes the functions of control unit 40 by executing programs loaded on RAM 1300. The HDD 1400 stores the data in the storage device of information processing device 10. The CPU 1100 of computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a predetermined communication network.

[0074] [8. Other] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0075] Furthermore, the configuration of the aforementioned information processing device 10 can be flexibly changed, for example, by calling external platforms, etc., via APIs (Application Programming Interfaces) or network computing, depending on the function.

[0076] Furthermore, the term "part" in the claims can be replaced with "means," "circuit," etc. For example, the estimation part can be replaced with estimation means or estimation circuit. [Explanation of symbols]

[0077] 1. Information Processing System 10 Information Processing Devices 20 Communications Department 30 Storage section 31 News Article Database 32 Comment Database 33 Model Databases 40 Control Unit 41 Estimation part 42 Generation part 43 Providing Department 100 User Terminals

Claims

1. An estimation unit that estimates the category of comments posted by users on news articles, Text information representing the comment in a style corresponding to the category estimated by the estimation unit, and a generation unit that generates text information aggregating the comment. An information processing device characterized by having the following features.

2. The estimation unit, The category of the aforementioned comment is estimated based on the context in which the comment is expressed. The information processing apparatus according to feature 1.

3. The estimation unit, The model, which has been trained to generate answers to input questions, is given the comment and an instruction sentence that instructs the model to output the category of the comment based on the context in which the comment is expressed, thereby estimating the category of the comment. The information processing apparatus according to feature 2.

4. The estimation unit, The category of the aforementioned comment is estimated based on information about the corresponding news article that corresponds to the aforementioned news article. The information processing apparatus according to feature 1.

5. The estimation unit, The category of a comment is estimated by inputting the comment, information about the corresponding news article, and an instruction sentence that instructs the model, which has been trained to generate answers to input questions, to output the category of the comment based on the information about the corresponding news article. The information processing apparatus according to feature 4.

6. The estimation unit, The system estimates whether the aforementioned comment belongs to the opinion category, which indicates the user's opinion, or the fact category, which indicates facts about the events described in the news article. The information processing apparatus according to feature 1.

7. The provisioning unit provides the text information to the user as a news article relating to the events described in the aforementioned news article. The information processing apparatus according to claim 1, further comprising the features.

8. The generating unit is The text information is generated by inputting the following to a model trained to generate answers to input questions: information about the comment, the category of the comment, and an instruction sentence that instructs the model to output the text information that represents the comment in a style corresponding to the category. The information processing apparatus according to feature 1.

9. A method of information processing performed by a computer, An estimation process to estimate the category of comments posted by users on news articles, Text information that shows the comments in a style corresponding to the category estimated by the estimation step, and a generation step that generates text information that aggregates the comments. An information processing method characterized by including

10. An estimation procedure for estimating the category of comments posted by users on news articles, Text information that shows the comments in a style corresponding to the category estimated by the estimation procedure, and a generation procedure that generates text information that aggregates the comments. An information processing program characterized by causing a computer to execute it.

Images