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

Abstract

This invention provides an information processing device, an information processing method, and an information processing program that appropriately generate articles from user comments. [Solution] In an information processing system in which a terminal device and an information processing device are connected to each other via a network by wired or wireless means so that they can communicate with each other, the control unit 130 of the information processing device 100 includes: an acquisition unit 131 that acquires multiple comments on a news article that is currently being discussed; a classification unit 132 that classifies the acquired multiple comments into multiple categories; a first generation unit 134 that generates multiple articles from the multiple comments classified into multiple categories; a determination unit 135 that compares the classification results of the multiple comments classified into multiple categories with the content of the multiple articles to determine whether or not there is a discrepancy in the content of the multiple articles; and a second generation unit 136 that, if there is a discrepancy in the content of the multiple articles, regenerates new articles for the articles in which the discrepancy has occurred.

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, there are online services that provide various contents such as news articles, columns, and product reviews to service users. In relation to such services, for example, technologies for automating news articles by selecting representative articles for each issue have also been proposed (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003] <000第00017号

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the conventional technology, there is room for improvement in appropriately generating articles from user comments.

[0005] The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program that can appropriately generate articles from user comments.

Means for Solving the Problems

[0006] The information processing device according to the present invention is characterized by comprising: an acquisition unit that acquires multiple comments on a news article that is currently being discussed; a classification unit that classifies the acquired multiple comments into multiple categories; a first generation unit that generates multiple articles from the multiple comments classified into the multiple categories; a determination unit that compares the classification results of the multiple comments classified into the multiple categories with the content of the multiple articles to determine whether or not there is a discrepancy in the content of the multiple articles; and a second generation unit that, if there is a discrepancy in the content of the multiple articles, newly generates articles for the articles in which the discrepancy has occurred. [Effects of the Invention]

[0007] According to one embodiment of the system, it is possible to appropriately generate articles from user comments. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 shows an example of the configuration of an information processing system according to an embodiment. [Figure 2] Figure 2 shows an example of information processing in the information processing system according to the embodiment. [Figure 3] Figure 3 shows an example of the configuration of a terminal device according to this embodiment. [Figure 4] Figure 4 shows an example of the configuration of an information processing device according to the present invention. [Figure 5] Figure 5 shows an example of a comment information storage unit according to the embodiment. [Figure 6] Figure 6 shows an example of a classification information storage unit according to the embodiment. [Figure 7] Figure 7 shows an example of a generated article information storage unit according to the embodiment. [Figure 8] Figure 8 is a flowchart showing the information processing procedure by the information processing system according to the embodiment. [Figure 9] Figure 9 is a hardware configuration diagram showing an example of a computer that implements the functions of an information processing device. [Modes for carrying out the invention]

[0009] 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.

[0010] (Embodiment) [1. Configuration of the Information Processing System] The information processing system 1 shown in Figure 1 will now be described. As shown in Figure 1, the information processing system 1 includes a terminal device 10 and an information processing device 100. The terminal device 10 and the information processing device 100 are connected to each other via a predetermined communication network (network N) by wired or wireless means. Figure 1 is a diagram showing an example configuration of the information processing system 1 according to an embodiment.

[0011] Terminal device 10 is an information processing device used by users who view articles via the internet. Users view articles (e.g., news articles) distributed through a designated web service, for example. Users also view (or attempt to view) comments posted by other readers of an article. Terminal device 10 can be any device as long as it can implement the processing described in the embodiment. Terminal device 10 may also be a smartphone, tablet, notebook PC, desktop PC, mobile phone, PDA, or other device. Figure 2 shows the case where terminal device 10 is a smartphone.

[0012] Terminal device 10 is, for example, a smart device such as a smartphone or tablet, and is a mobile terminal device that can communicate with any server device via a wireless communication network such as 4G-5G (Generation) or LTE (Long Term Evolution). Terminal device 10 also has a screen, such as an LCD display, which has touch panel functionality and may accept various operations on displayed data such as content from the user, such as tapping, sliding, and scrolling, using a finger or stylus. In Figure 2, terminal device 10 is used by user U1.

[0013] The information processing device 100 is an information processing device intended to appropriately generate articles from user comments, and can be any device as long as it can realize the processing in the embodiment. The information processing device 100 can be realized by, for example, a server device or cloud system that provides a predetermined article that the user is viewing, or a server device or cloud system that provides a predetermined web service to which such predetermined article is distributed.

[0014] The information processing device 100, for example, acquires comments posted on a predetermined article, generates an article, and provides the generated article to the terminal device 10. Specifically, the information processing device 100 acquires comments on a news article that is currently a hot topic and classifies the acquired comments. The information processing device 100 places the classified comments in a placement area, generates multiple articles from the placed comments, and selects one article. The information processing device 100 then provides the selected article to the terminal device 10. In this way, the information processing device 100 can appropriately generate articles from user comments.

[0015] In addition, the information processing apparatus 100 has a model. The model is, for example, a text generation model trained with data published on the Internet or the like. For example, the model may be implemented by an AI (Artificial Intelligence) such as GPT (Generative Pre-trained Transformer). GPT is a text generation AI and a language model capable of generating text using natural language processing.

[0016] In FIG. 1, the case where the terminal device 10 and the information processing apparatus 100 are separate devices is shown, but the terminal device 10 and the information processing apparatus 100 may be integrated.

[0017] 〔2. An Example of Information Processing〕 〔2.1. Basic Processing〕 FIG. 2 is a diagram showing an example of information processing of the information processing system 1 according to the embodiment.

[0018] First, the information processing apparatus 100 acquires a comment C (step S1). The information processing apparatus 100 acquires a plurality of comments C1 to C7 on a news article that is the topic. When acquiring a comment, the information processing apparatus 100 acquires the article to which the comment was posted. It is preferable for the information processing apparatus 100 to acquire comments posted on each article from a plurality of news articles related to the topic, but comments on one news article may be acquired. In the present embodiment, the topic is the championship competition of sports (for example, a baseball team) as an example, but any topic may be used.

[0019] Next, the information processing device 100 classifies the comments (step S2). The information processing device 100 classifies the acquired comments C. The information processing device 100 classifies the acquired comments C1 to C7. For example, the information processing device 100 uses natural language processing technology to classify the acquired comments C1 to C7 into multiple categories. The information processing device 100 uses natural language processing technology to analyze the frequency of keyword occurrences in the acquired comments C1 to C7 and perform sentiment analysis from the content of the acquired comments to classify them into multiple categories. The information processing device 100 uses natural language processing technology to analyze the content of the acquired comments C1 to C7 and classify the analyzed comments C1 to C7 into three categories. For example, the information processing device 100 uses natural language processing technology to analyze the content of the acquired comments C1 to C7 and classify the analyzed comments C1 to C7 into three categories: majority, minority, and neutral. Keywords are expressions that indicate the core of the article or comment in which the comment was posted, such as "AA" or "champion."

[0020] In this embodiment, there are three categories: majority, minority, and moderate. However, depending on the content of the article and comments, there may be two or more categories. Furthermore, the categories may be pre-set, or they may be generated by the information processing device 100 when classifying acquired comments, based on the content of the article and comments. Also, the content of the categories is not limited to majority, minority, moderate, etc., and may be set according to the content of the article and comments.

[0021] In the example shown in Figure 2, the information processing device 100 classifies the acquired comments C1 to C7 into these opinions. The information processing device 100 classifies the acquired comments C1, C5, and C6 as majority opinions, comment C2 as a minority opinion, and C3, C4, and C7 as neutral opinions.

[0022] The information processing device 100 places the classified comments into their respective placement areas (step S3). The information processing device 100 uses a generation AI to place the classified comments C1 to C7 into their respective placement areas SS. For example, the information processing device 100 uses a GPT model to place the classified comments C1 to C7 into their respective placement areas SS. The information processing device 100 causes the generation AI to place the classified comments into each area of ​​the placement area SS to generate a placement pattern for the placement area SS. The information processing device 100 causes the generation AI to place the classified comments into each area of ​​the placement area SS to generate a placement pattern for each area of ​​the placement area SS.

[0023] The placement area SS is, for example, an area where classified comments are appropriately placed, such as a sweet spot. The placement area SS is divided into categories into which multiple comments are classified. The placement area SS may be implemented as a table for placing classified comments. In this embodiment, the placement area SS is implemented as, for example, a single table, which is divided into a higher-level area AR1 for placing majority opinions, a lower-level area AR2 for placing minority opinions, and an intermediate area AR3 for placing opinions in between. The intermediate area AR3 is located between the higher-level area AR1 and the lower-level area AR2. Note that the placement area SS is not limited to having three areas within a single table; a separate table may be provided for each category.

[0024] In Figure 2, the information processing device 100 has the generating AI place comments C1, C5, and C6, which are classified as majority, in the upper region AR1 of the placement region SS; comment C2, which is classified as minority, in the lower region AR2 of the placement region SS; and comments C3, C4, and C7, which are classified as neutral, in the intermediate region AR3 between the upper region AR1 and the lower region AR2 of the placement region. For example, the information processing device 100 has the generating AI place multiple comments classified based on the content of multiple comments and evaluations of multiple comments into each region of the placement region SS. Evaluations of multiple comments include the number of "likes" and shares for each comment.

[0025] Next, the information processing device 100 generates multiple articles from the classified comments (step S4). The information processing device 100 generates multiple articles from the comments placed in the placement area SS. In this case, the information processing device 100 generates multiple articles by using a generation AI. For example, the information processing device 100 generates multiple articles by using a GPT model. The multiple articles may be news articles or survey articles, but they are of the same type and concern the same event. For example, in Figure 2, the information processing device 100 shows multiple news articles about the baseball championship race. It is also preferable that the number of articles generated is three. In other words, in Figure 2, the information processing device 100 generates three news articles about the baseball championship race.

[0026] Here, we will describe the GPT model according to the embodiment. The GPT model according to the embodiment, for example, when prompted with a comment to be acquired and an instruction to place the acquired comment in a placement area SS, uses the prompts to place the comment in the placement area SS. Alternatively, the GPT model according to the embodiment, when prompted with a comment placed in the placement area SS and an instruction to generate multiple articles, uses the prompts to generate articles.

[0027] An example of a prompt for placing multiple classified comments in their respective placement areas SS is: "Place the classified comments in their placement areas. Place majority opinions in the upper area of ​​the placement area, minority opinions in the lower area of ​​the placement area, and moderate opinions in the intermediate area between the upper and lower areas of the placement area. If the classified comments have evaluations such as the number of 'likes,' the number of shares, or replies to the comments, it is preferable to also consider the evaluations of the comments. The following are the acquired comments." For example, the GPT model places comments that are more assertive in the upper areas of each region. The GPT model may also place comments that avoid assertions in the upper areas of each region.

[0028] Furthermore, an example of a prompt for generating multiple articles from placed comments is: "Generate three articles of the same type from the placed comments. This time, generate three news articles. The articles should include the majority, minority, and moderate opinions of the classified comments. Also, generate the articles in a way that avoids definitive statements. In addition, generate the articles with a consistent writing style. Also, to generate the articles accurately, refer to facts and reliable information in addition to the comments." For example, the GPT model will determine that the more frequently a keyword is included in the placed comments, the more important that keyword is and will generate articles accordingly. For example, the GPT model will determine that the keyword is important if the same keyword is included multiple times in the placed comments and will generate articles accordingly.

[0029] Next, the information processing device 100 compares the classification results of multiple comments classified into multiple categories with the content of multiple articles to determine whether or not there is a discrepancy in the content of the multiple articles (step S4). The information processing device 100 compares the content of comments placed in the placement area SS with the content of the generated multiple articles to determine whether or not there is a discrepancy in the content of each article. The information processing device 100 determines whether or not there is a discrepancy in the content of the multiple articles based on whether or not the content of the multiple articles contains multiple categories.

[0030] For example, the information processing device 100 determines whether the content of multiple articles contains comments classified into three categories. The information processing device 100 determines whether the comments contain comments classified into three categories: majority, minority, and moderate. If the information processing device 100 determines that the content of multiple articles contains comments classified into all categories, it determines that there is no discrepancy in the content of the multiple articles. If the content of multiple articles does not contain at least one comment classified into all categories, it determines that there is a discrepancy in the content of the multiple articles.

[0031] Next, if there are discrepancies in the content of multiple articles, the information processing device 100 regenerates the articles (step S5). If there are discrepancies in the content of the articles, the information processing device 100 regenerates the articles in which discrepancies have occurred. When the information processing device 100 regenerates an article, it updates the old article with the new article. For example, in Figure 2, the information processing device 100 regenerates article A1a, which includes minority and moderate opinions, because article A1 only contains the majority opinion. The information processing device 100 updates the old article A1 with the regenerated article A1a. If there are discrepancies in the content of all three generated articles A1 to A3, the information processing device 100 regenerates all articles and updates all the old articles with the newly generated articles.

[0032] Next, the information processing device 100 selects one article from the multiple articles generated (step S6). If there are no discrepancies in the content of the multiple articles, the information processing device 100 selects one article from the multiple articles based on selection criteria. Multiple selection criteria are set, and examples include whether the prompt for generating multiple articles has been followed, whether there are any expressions that may cause misunderstanding, whether the information is accurate, whether the writing style of the articles is consistent, and whether the content is of interest to the user viewing the article. For example, the information processing device 100 determines whether the multiple articles A1a to A3 satisfy these selection criteria and selects one article that satisfies the selection criteria. The specific article selection process will be described later.

[0033] Next, the information processing device 100 provides the selected article to the terminal device 10 (step S7). In Figure 2, the information processing device 100 provides article A2 to the terminal device 10. The terminal device 10 displays the article provided by the information processing device 100.

[0034] [2.2. Adjusting comments placed in the layout area] The information processing device 100 adjusts the comments placed in the placement area SS. The information processing device 100 causes the generating AI to dynamically adjust the placement of the comments in the placement area SS. For example, the information processing device 100 causes the generating AI to dynamically adjust the comments placed in each area of ​​the placement area SS if a discrepancy occurs in the placement of the classified comments in each area of ​​the placement area SS. A discrepancy in the placement of comments may occur, for example, when the comments are not placed correctly.

[0035] For example, the information processing device 100 may also adjust the placement of comments in real time. The information processing device 100 acquires and classifies comments in real time whenever comments are posted on a news article that is currently being discussed. The information processing device 100 then causes the generating AI to dynamically adjust the placement of comments in the placement area SS, taking into account the comments that have been acquired and classified in real time.

[0036] Alternatively, for example, the information processing device 100 may instruct the generation AI to dynamically adjust the placement of comments based on the content of the multiple articles that have been generated. The information processing device 100 may also instruct the generation AI to learn from the multiple articles that have been generated and the placement of comments (placement patterns) in the placement area SS at the time those articles were generated, and then dynamically adjust the placement area SS by reflecting the learning results in its generation.

[0037] Dynamic adjustment of comment placement includes, for example, rearranging the order of placed comments, replacing placed comments with new comments, adding new comments, rearranging incorrectly placed comments, and deleting comments that have already been placed. The information processing device 100 dynamically adjusts the comment placement by performing at least one of these actions.

[0038] For example, the information processing device 100 may have the generation AI analyze the generation results of past articles according to the theme of the article, select the arrangement pattern of the arrangement areas when past articles were generated, and adjust the arrangement of comments placed in the arrangement areas. When generating the next article, the information processing device 100 may have the generation AI obtain the arrangement areas SS that were adjusted in the past according to the theme of the article, and place comments in the arrangement areas when generating the next article, taking into account the arrangement areas SS that were adjusted in the past. For example, the information processing device 100 may evaluate the generated article and obtain the arrangement pattern of the arrangement areas SS that maximized the value of the article. The information processing device 100 may evaluate the generated article and obtain the arrangement pattern of comments in the arrangement areas SS where there was no discrepancy in the content of the generated article, and reflect this in the generation of the arrangement areas when generating the next article. The generation of the next article here refers to the generation of multiple articles in step 4.

[0039] [2.3. Article Selection Process] The information processing device 100 determines whether the prompt for generating multiple articles has been followed and selects one article from the multiple articles A1a to A3. For example, the information processing device 100 determines whether the generated multiple articles A1a to A3 are neutral in their expression based on facts and selects one article from the multiple articles A1a to A3.

[0040] The information processing device 100 determines whether the content of multiple articles contains any misleading expressions. The information processing device 100 determines whether the classified comments and facts are distinguishable. The information processing device 100 determines whether the information used when generating multiple articles is accurate. When generating an article from comments placed in the placement area SS, the information processing device 100 refers to content related to the article. In this embodiment, content related to the article includes the championship history of the championship match between AA and BB. In the example shown in Figure 2, the selected article A2 indicates that BB's victory is a fact by using expressions such as "BB also has a history of winning the championship."

[0041] The information processing device 100 determines whether the writing style of multiple articles is consistent. The information processing device 100 also determines whether readability is maintained. For example, the information processing device 100 determines whether formal or colloquial sentences are mixed in with articles written in the plain style.

[0042] The information processing device 100 determines whether multiple articles will attract the user's interest. The information processing device 100 determines whether the content of multiple articles will attract the target reader's interest. For example, the information processing device 100 determines whether an article will attract the reader's interest based on the topic of comments or trending themes.

[0043] In Figure 2, the information processing device 100 selects article A2, which complies with the prompt.

[0044] The information processing device 100 may regenerate articles where there is a misalignment, based on the arrangement area SS in which the comments have been adjusted.

[0045] In the above example, three articles are generated for a single theme, but multiple articles may be generated for a category. When generating multiple articles for a category, the information processing device 100 may generate articles of different genres. For example, if the category is sports, the information processing device 100 may generate articles about soccer, baseball, and tennis. In this case, the information processing device 100 retrieves comments for each genre of article and generates multiple articles.

[0046] [3. Configuration of the terminal device] Next, the configuration of the terminal device 10 according to the embodiment will be described using Figure 3. Figure 3 is a diagram showing an example of the configuration of the terminal device 10 according to the embodiment. As shown in Figure 3, the terminal device 10 has a communication unit 11, an input unit 12, an output unit 13, and a control unit 14.

[0047] (Communications Section 11) The communication unit 11 is implemented, for example, by a NIC (Network Interface Card). The communication unit 11 is connected to a predetermined network N by wire or wireless connection and sends and receives information to and from the information processing device 100 via the predetermined network N.

[0048] (Input section 12) The input unit 12 receives various operations from the user. In Figure 2, it receives various operations from user U1. For example, the input unit 12 may receive various operations from the user via the display surface using a touch panel function. Alternatively, the input unit 12 may receive various operations from buttons provided on the terminal device 10, or from a keyboard or mouse connected to the terminal device 10.

[0049] (Output section 13) The output unit 13 is a display screen for a tablet terminal, for example, which is implemented using a liquid crystal display or an organic EL (Electro-Luminescence) display, and is a display device for displaying various types of information. For example, the output unit 13 displays content based on information transmitted from the information processing device 100.

[0050] (Control Unit 14) The control unit 14 is, for example, a controller, and is implemented by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs stored in the internal memory of the terminal device 10 using RAM (Random Access Memory) as the working area. For example, these various programs include application programs installed on the terminal device 10. For example, these various programs include application programs that display content based on information transmitted from the information processing device 100. The control unit 14 is also implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

[0051] As shown in Figure 3, the control unit 14 has a receiving unit 141 and a transmitting unit 142, and realizes or executes the information processing operations described below.

[0052] (Receiver 141) The receiving unit 141 receives information transmitted from, for example, the information processing device 100. For example, the receiving unit 141 receives information such as a predetermined article or voting content that the user is viewing as feedback on the user's operation.

[0053] (Transmitter 142) The transmission unit 142 transmits, for example, user operation information to the terminal device 10. For example, the transmission unit 142 transmits operation information performed by the user on the UI screen of a comment posted on a predetermined article that the user is viewing. For example, the transmission unit 142 transmits operation information such as posting a comment on an article or voting information from voting content.

[0054] [4. Configuration of the Information Processing Device] Next, the configuration of the information processing device 100 according to the embodiment will be described using Figure 4. Figure 4 is a diagram showing an example of the configuration of the information processing device 100 according to the embodiment. As shown in Figure 4, the information processing device 100 has a communication unit 110, a storage unit 120, and a control unit 130. The information processing device 100 may also have an input unit (for example, a keyboard or mouse) that receives various operations from the administrator of the information processing device 100, and a display unit (for example, a liquid crystal display) for displaying various information.

[0055] (Communications Department 110) The communication unit 110 is implemented, for example, by a NIC. The communication unit 110 is connected to the network N by wire or wireless connection and sends and receives information to and from terminal devices 10, etc., via the network N.

[0056] (Storage unit 120) The storage unit 120 is implemented by, for example, semiconductor memory elements such as RAM and flash memory, or storage devices such as hard disks and optical discs. As shown in Figure 4, the storage unit 120 has a comment information storage unit 121, a classification information storage unit 122, and a generated article information storage unit 123.

[0057] The comment information storage unit 121 stores information about the acquired comments. For example, the comment information storage unit 121 stores information about the comment content and posting date and time. Here, Figure 5 shows an example of the comment information storage unit 121 according to the embodiment. Figure 5 is a diagram showing an example of the comment information storage unit according to the embodiment. As shown in Figure 5, the comment information storage unit 121 has items such as "comment ID", "comment content", and "posting date and time".

[0058] The "Comment ID" indicates identification information used to identify a comment. The "Comment Content" indicates the content of the comment. In the example shown in Figure 5, conceptual information such as "Comment Content #1" and "Comment Content #2" is stored in "Comment Content," but in reality, the text of the article is stored there. The "Posting Date and Time" indicates the posting date and time the comment was posted. In the example shown in Figure 5, conceptual information such as "Posting Date and Time #1" and "Posting Date and Time #2" is stored there, but in reality, the date and time the comment was posted is stored there.

[0059] The classification information storage unit 122 stores information related to classification. For example, the classification information storage unit 122 stores information related to classification content and the number of comments. Here, Figure 6 shows an example of the classification information storage unit 122 according to the embodiment. Figure 6 is a diagram showing an example of the classification information storage unit according to the embodiment. As shown in Figure 6, the classification information storage unit 122 has items such as "classification ID", "classification content", and "number of comments".

[0060] The "Classification ID" indicates identification information used to identify a classification. The "Classification Content" indicates the content of the classification. In the example shown in Figure 6, conceptual information such as "Classification Content #1" and "Classification Content #2" is stored in "Classification Content," but in reality, the content of the classification (e.g., majority) is stored there. The "Number of Comments" indicates the number of comments classified under the classification content. In the example shown in Figure 6, conceptual information such as "Number of Comments #1" and "Number of Comments #2" is stored in "Number of Comments," but in reality, the number of comments classified under the classification content is stored there.

[0061] The generated article information storage unit 123 stores information about the generated article. For example, the generated article information storage unit 123 stores information such as the article's content, title, category, and delivery date and time. Here, Figure 7 shows an example of the generated article information storage unit 123 according to the embodiment. Figure 7 is a diagram showing an example of the generated article information storage unit according to the embodiment. As shown in Figure 7, the generated article information storage unit 123 further has items such as "generated article ID", "content", "title", "category", and "delivery date and time".

[0062] "Generated Article ID" indicates identification information for identifying the article. "Content" indicates the content of the article. In the example shown in Figure 7, conceptual information such as "Content #1" and "Content #2" is shown as being stored in "Content," but in reality, the text of the article is stored there. "Title" indicates the title of the article. In the example shown in Figure 7, conceptual information such as "Title #1" and "Title #2" is shown as being stored in "Title," but in reality, the text of the title is stored there. "Category" indicates the category of the article (for example, sports, entertainment, incidents and accidents, etc.). The category may also be, for example, controversial, royal family, politics, international, domestic, LGBT, lifestyle, etc. "Distribution Date and Time" indicates the distribution date and time of the article.

[0063] (Control unit 130) The control unit 130 is a controller, and is implemented, for example, by a CPU or MPU executing various programs stored in the memory device inside the information processing device 100 using RAM as the working area. Alternatively, the control unit 130 can be implemented by an integrated circuit such as an ASIC or FPGA.

[0064] As shown in Figure 4, the control unit 130 includes an acquisition unit 131, a classification unit 132, an arrangement unit 133, a first generation unit 134, a determination unit 135, a second generation unit 136, a selection unit 137, and a provision unit 138, and realizes or executes the information processing operations described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in Figure 4, and other configurations are also acceptable as long as they perform the information processing described later.

[0065] (Acquisition part 131) The acquisition unit 131 acquires various information from the storage unit 120. The acquisition unit 131 also stores the acquired information in the storage unit 120.

[0066] The acquisition unit 131 acquires various information from external information processing devices. The acquisition unit 131 also acquires various information from other information processing devices such as the terminal device 10.

[0067] For example, the acquisition unit 131 acquires comments. The acquisition unit 131 acquires multiple comments C (comments C1 to C7 in this embodiment) for a news article that is the topic of discussion. In addition, when acquiring comments, the acquisition unit 131 also acquires the news article in which the acquired comments were posted.

[0068] Furthermore, the acquisition unit 131 acquires comments on trending news articles in real time. For example, the acquisition unit 131 acquires comments in real time whenever comments are posted on a trending news article.

[0069] (Classification section 132) The classification unit 132 classifies the acquired comments C. The classification unit 132 classifies the multiple comments C1 to C7 acquired by the acquisition unit 131 into multiple categories. The classification unit 132 classifies the multiple comments C1 to C7 acquired into multiple categories using, for example, natural language processing technology. For example, the classification unit 132 classifies the multiple comments C1 to C7 acquired into multiple categories using natural language processing technology. The classification unit 132 uses natural language processing technology to analyze the frequency of keyword occurrences in the multiple comments C1 to C7 acquired and to perform sentiment analysis from the content of the multiple comments C1 to C7 acquired, and classifies them into multiple categories. The classification unit 132 uses natural language processing technology to analyze the content of the multiple comments C1 to C7 acquired and classifies the analyzed multiple comments C1 to C7 into three categories.

[0070] Specifically, the classification unit 132 uses natural language processing technology to analyze the content of multiple comments C1 to C7 that it has acquired, and classifies the analyzed comments C1 to C7 into three categories: majority, minority, and neutral.

[0071] Furthermore, the classification unit 132 classifies the acquired comments in real time. For example, if the acquisition unit 131 acquires comments on a trending news article in real time, the classification unit 132 classifies the acquired comments in real time.

[0072] (Arrangement part 133) The placement unit 133 places the classified comments C1 to C7 into the placement area SS. The placement unit 133 uses a generation AI to place the classified comments C1 to C7 into the placement area SS. For example, the placement unit 133 uses a GPT model to place the classified comments C1 to C7 into the placement area SS. The placement unit 133 has the generation AI place the classified comments into the placement area SS to generate the placement area SS.

[0073] The placement unit 133 causes the generation AI to place the classified comments in each area of ​​the placement area SS based on the content of the multiple comments C1 to C7 and the evaluations of the multiple comments C1 to C7. For example, the placement unit 133 causes the generation AI to place comments classified as majority in the upper area AR1 of the placement area SS, comments classified as minority in the lower area AR2 of the placement area SS, and comments classified as neutral in the intermediate area AR3 between the upper area AR1 and the lower area AR2 of the placement area SS.

[0074] The placement unit 133 adjusts the comments placed in the placement area SS. The placement unit 133 causes the generation AI to dynamically adjust the placement of the comments placed in the placement area SS. For example, if the generation AI has placed the classified comments in each area of ​​the placement area SS and a discrepancy has occurred in the placement of the comments, the placement unit 133 causes the generation AI to dynamically adjust the comments placed in each area of ​​the placement area SS.

[0075] Alternatively, for example, the placement unit 133 may adjust the placement of comments in real time. The placement unit 133 causes the generation AI to dynamically adjust the comments placed in the placement area SS in real time.

[0076] Alternatively, for example, the placement unit 133 may cause the generation AI to dynamically adjust the placement of comments based on the content of the multiple generated articles. The placement unit 133 may also cause the generation AI to learn by inputting the multiple generated articles and the placement (placement pattern) of comments in the placement area SS at the time the articles were generated, and then dynamically adjust the placement area SS by reflecting the learning results in its generation.

[0077] Dynamic adjustment of comment placement includes, for example, rearranging the order of placed comments, replacing placed comments with new comments, adding new comments, rearranging incorrectly placed comments, and deleting comments that have already been placed. The placement unit 133 dynamically adjusts the placement of comments by performing at least one of these actions.

[0078] The placement unit 133 instructs the generation AI to analyze the generation results of past articles according to the theme of the article, select the placement pattern of the placement area when the past article was generated, and adjust the placement of comments placed in the placement area. In other words, when generating the next article, the placement unit 133 may instruct the generation AI to obtain the previously adjusted placement area SS according to the theme of the article, and to place comments in the placement area when generating the next article, taking into account the previously adjusted placement area SS. For example, the placement unit 133 may evaluate the generated article and obtain the placement pattern of the placement area SS that maximizes the value of the article. The placement unit 133 may also evaluate the generated article and obtain the placement pattern of comments in the placement area SS where there was no discrepancy in the content of the generated article, and reflect this in the generation of the placement area when generating the next article.

[0079] (first generation unit 134) The first generation unit 134 generates multiple articles from multiple classified comments, for example. The first generation unit 134 generates multiple articles from multiple comments classified into multiple categories by using a generation AI. For example, the first generation unit 134 generates multiple articles from multiple classified comments by using a GPT model. The first generation unit 134 generates multiple articles of the same type. For example, the first generation unit 134 generates three news articles.

[0080] (Judgment unit 135) The determination unit 135 compares the classification results of multiple comments classified into multiple categories with the content of multiple articles to determine whether there is a discrepancy in the content of the multiple articles. The determination unit 135 determines whether there is a discrepancy in the content of multiple articles by checking whether the content of multiple articles contains the opinions of comments classified into multiple categories. The determination unit 135 compares the content of multiple comments placed in the placement area SS by the placement unit 133 with the content of multiple articles to determine whether there is a discrepancy in the content of each article.

[0081] For example, the determination unit 135 determines whether the content of multiple articles includes comments classified into three categories. The determination unit 135 determines whether the comments include comments classified into three categories: majority, minority, and moderate. If the determination unit 135 determines that the content of multiple articles includes comments classified into all categories, it determines that there is no discrepancy in the content of the multiple articles. If the content of multiple articles does not include at least one comment classified into all categories, the determination unit 135 determines that there is a discrepancy in the content of the multiple articles.

[0082] (Second generation unit 136) The second generation unit 136 regenerates the articles if the determination unit 135 determines that there are discrepancies in the content of multiple articles. If there are discrepancies in the content of multiple articles, the second generation unit 136 regenerates new articles for the articles that have discrepancies. If the second generation unit 136 regenerates articles, it updates the old articles with the new articles.

[0083] Furthermore, if there are discrepancies in the content of all three generated articles A1 to A3, the second generation unit 136 may regenerate all articles and update all the old articles with the regenerated articles.

[0084] Furthermore, the second generation unit 136 may regenerate articles where there is a misalignment, based on the arrangement area SS in which the comments have been adjusted. In this case, the arrangement unit 133 adjusts the comments placed in the arrangement area SS.

[0085] (Selection section 137) The selection unit 137 selects one article from the multiple articles generated. If there are no discrepancies in the content of the multiple articles, the selection unit 137 selects one article from the multiple articles based on the selection criteria. For example, the selection unit 137 determines whether the selection criteria are met for multiple articles A1a to A3, and selects one article that meets the selection criteria.

[0086] For example, the selection unit 137 determines whether the prompt for generating multiple articles has been followed and selects one article from the multiple articles A1a to A3. The selection unit 137 then determines, for example, whether the generated multiple articles A1a to A3 are neutral in their expression based on facts and selects one article from the multiple articles A1a to A3.

[0087] For example, the selection unit 137 determines whether the content of the generated articles contains any misleading expressions. The selection unit 137 also determines whether the content of the classified comments is distinguishable from the facts. Furthermore, for example, the selection unit 137 determines whether the information used when generating the articles is accurate. When generating articles from comments placed in the placement area SS, the selection unit 137 refers to content related to the articles.

[0088] The selection unit 137 determines whether the writing style of multiple articles is consistent. The selection unit 137 also determines whether readability is maintained. For example, the selection unit 137 determines whether formal or colloquial sentences are mixed in with articles written in the plain style.

[0089] The selection unit 137 determines whether the content of multiple articles will attract the user's interest. The selection unit 137 determines whether the content of multiple articles will attract the user's interest when viewing the articles. For example, the selection unit 137 determines whether the content will attract the user's interest based on the topic of comments or trending themes.

[0090] (Provider 138) The provisioning unit 138 provides the article selected by the selection unit 137 to the terminal device 10.

[0091] [5. Information Processing Flow] Next, the information processing procedure by the information processing system 1 according to the embodiment will be explained using Figure 8. Figure 8 is a flowchart showing the information processing procedure by the information processing system 1 according to the embodiment.

[0092] As shown in Figure 6, the information processing device 100 obtains comments on the news article (step S100).

[0093] The information processing device 100 classifies the acquired comments (step S101).

[0094] The information processing device 100 places the classified comments (step S102). The information processing device 100 places the comments in the placement area SS.

[0095] The information processing device 100 generates multiple articles based on the placed comments (step S103).

[0096] The information processing device 100 determines whether or not there is a discrepancy in the content of the multiple articles generated (step S104). The information processing device 100 compares the classification results of the classified comments with the content of the first multiple articles generated to determine whether or not there is a discrepancy.

[0097] If the information processing device 100 determines that there is a discrepancy in the content of multiple articles (step S104; Yes), it plays back the articles that have discrepancies (step S105) and proceeds to step S106. If the information processing device 100 determines that there is no discrepancy in the content of multiple articles (step S104; No), it selects one article based on the selection criteria (step S106) and provides the selected article to the terminal device 10 (step S107).

[0098] [6. Effects] As described above, the information processing device 100 according to the embodiment is characterized by comprising: an acquisition unit 131 that acquires multiple comments on a news article that is currently being discussed; a classification unit 132 that classifies the acquired multiple comments into multiple categories; a first generation unit 134 that generates multiple articles from the multiple comments classified into multiple categories; a determination unit 135 that compares the classification results of the multiple comments classified into multiple categories with the content of the multiple articles to determine whether or not there is a discrepancy in the content of the multiple articles; and a second generation unit 136 that, if there is a discrepancy in the content of the multiple articles, newly generates articles for the articles in which the discrepancy has occurred.

[0099] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0100] Furthermore, the information processing device 100 includes a placement unit 133 that uses a generation AI to place the classified comments in their respective placement areas, and a selection unit 137 that, if there is no discrepancy in the content of multiple articles, selects one article from among the multiple articles based on selection criteria.

[0101] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments and select appropriate articles.

[0102] The classification unit 132 uses natural language processing technology to analyze the content of multiple acquired comments and classifies the analyzed comments into categories of majority, minority, and neutral. The placement unit 133 instructs the generation AI to place comments classified as majority in the upper-level area of ​​the placement domain, comments classified as minority in the lower-level area of ​​the placement domain, and comments classified as neutral in the intermediate area between the upper-level and lower-level areas of the placement domain.

[0103] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0104] Furthermore, the classification unit 132 uses natural language processing technology to analyze the frequency of keyword occurrences in the acquired comments and to perform sentiment analysis based on the content of the acquired comments.

[0105] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0106] Furthermore, the placement unit 133 causes the generating AI to place the classified comments into each area of ​​the placement area based on the content of the multiple comments and the evaluation of the multiple comments.

[0107] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0108] Furthermore, the placement unit 133 causes the generating AI to dynamically adjust the comments placed in the placement area SS.

[0109] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0110] Furthermore, the acquisition unit 131 acquires comments on trending news articles in real time. When the acquisition unit 131 acquires comments in real time, the classification unit 132 classifies the acquired comments in real time. The placement unit 133 causes the generation AI to dynamically adjust the placement of the comments placed in the placement area SS in real time.

[0111] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0112] Furthermore, the placement unit 133 causes the generation AI to analyze the generation results of past articles according to the theme of the article, select the placement pattern of the placement area when generating past articles, and adjust the placement of the comments placed in the placement area.

[0113] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0114] The selection unit 137 determines whether the prompts for generating multiple articles have been followed for each of the multiple articles.

[0115] As a result, the information processing device 100 according to this embodiment can appropriately generate articles from user comments.

[0116] Furthermore, the selection unit 137 determines whether the classified comments and facts are distinguishable for each of the articles.

[0117] This allows the information processing device 100 according to the embodiment to suppress misunderstandings for the reader.

[0118] Furthermore, the selection unit 137 determines whether the information used when generating multiple articles is accurate.

[0119] As a result, the information processing device 100 according to the embodiment can suppress confusion and misunderstanding among users.

[0120] Furthermore, the selection unit 137 determines whether the writing style of multiple articles is consistent or not.

[0121] As a result, the information processing device 100 according to this embodiment can select articles that are easy for the user to read.

[0122] Furthermore, the selection unit 137 determines whether multiple articles are of interest to the user viewing the articles.

[0123] As a result, the information processing device 100 according to the embodiment can select articles that are of interest to the user.

[0124] Furthermore, the determination unit 135 determines whether there is a discrepancy in the content of the multiple articles by checking whether the content of the multiple articles includes comments classified into the multiple categories.

[0125] As a result, the information processing device 100 according to the embodiment can determine whether an article contains a variety of opinions from users.

[0126] Furthermore, if the determination unit 135 determines that the content of multiple articles includes comments classified into multiple categories, it determines that there is no discrepancy in the content of the multiple articles.

[0127] As a result, the information processing device 100 according to the embodiment can determine whether an article contains a variety of opinions from users.

[0128] [7. Hardware Configuration] Furthermore, the information processing device 100 according to the above-described embodiment is realized by a computer 1000 having the configuration shown in Figure 9. Figure 9 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device 100. The computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0129] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls various parts. The ROM 1300 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.

[0130] The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 acquires data from other devices via a predetermined communication network and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the predetermined communication network.

[0131] 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 generated data to output devices via the input / output interface 1600.

[0132] 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 can be, 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), tape media, magnetic recording medium, or semiconductor memory.

[0133] For example, when the computer 1000 functions as an information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 130 by executing a program loaded on the RAM 1200. The CPU 1100 of the 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.

[0134] [8. Other] 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 known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document 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.

[0135] 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.

[0136] Furthermore, the embodiments described above can be combined as appropriate, as long as the processing content is not contradictory.

[0137] 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.

[0138] Furthermore, the terms "section, module, unit" mentioned above can be replaced with "means" or "circuit," etc. For example, the acquisition unit can be replaced with acquisition means or acquisition circuit. [Explanation of Symbols]

[0139] 1. Information Processing System 10 Terminal devices 11 Communications Department 12 Input section 13 Output section 14 Control Unit 100 Information Processing Devices 110 Communications Department 120 Storage section 121 Comment Information Storage Unit 122 Classification information storage unit 123 Generated Article Information Storage Unit 130 Control Unit 131 Acquisition Department 132 Classification Department 133 Placement section 134 1st generation part 135 Judgment section 136 Second generation part 137 Selection Section 138 Provision Department N Network

Claims

1. A unit that retrieves multiple comments on a trending news article, A classification unit that classifies the acquired multiple comments into multiple categories, A first generation unit that generates multiple articles from the multiple comments classified into the multiple categories, A determination unit that compares the classification results of the multiple comments classified into the multiple categories with the content of the multiple articles to determine whether or not there is a discrepancy in the content of the multiple articles, If the aforementioned discrepancies occur in the content of the multiple articles, a second generation unit regenerates the articles in which the discrepancies occur, An information processing device characterized by having the following features.

2. A placement unit that uses generation AI to place the classified multiple comments into their respective placement areas, If there is no discrepancy in the content of the aforementioned multiple articles, the system further includes a selection unit that selects one article from the aforementioned multiple articles based on selection criteria. The information processing apparatus according to feature 1.

3. The classification unit uses natural language processing technology to analyze the content of the acquired multiple comments and classifies the analyzed multiple comments into categories of majority, minority, and neutral. The arrangement unit causes the generating AI to arrange comments classified as majority in the upper area of ​​the arrangement region, comments classified as minority in the lower area of ​​the arrangement region, and comments classified as neutral in the intermediate area between the upper and lower areas of the arrangement region. The information processing apparatus according to feature 2.

4. The classification unit uses the natural language processing technology to analyze the frequency of occurrence of keywords in the acquired comments and to perform sentiment analysis on the content of the acquired comments. The information processing apparatus according to claim 3.

5. The arrangement unit causes the generation AI to arrange the classified comments in each area of ​​the arrangement region based on the content of the comments and the evaluation of the comments. The information processing apparatus according to claim 3.

6. The arrangement unit causes the generating AI to dynamically adjust the comments placed in the arrangement area. The information processing apparatus according to claim 3.

7. The aforementioned acquisition unit acquires comments on the news article in question in real time, When the acquisition unit acquires comments in real time, the classification unit classifies the acquired comments in real time. The arrangement unit causes the generation AI to dynamically adjust the arrangement of comments placed in the arrangement area in real time. The information processing apparatus according to feature 6.

8. The arrangement unit causes the generation AI to analyze the generation results of past articles according to the theme of the article, to select the arrangement pattern of the arrangement area used when the past article was generated, and to adjust the arrangement of the comments placed in the arrangement area. The information processing apparatus according to claim 3.

9. The selection unit determines whether or not the prompts for generating the multiple articles have been followed. The information processing apparatus according to feature 2.

10. The selection unit determines whether the content of the classified comments is distinguishable from the facts. The information processing apparatus according to feature 2.

11. The selection unit determines whether the information used to generate the multiple articles is accurate. The information processing apparatus according to feature 2.

12. The selection unit determines whether the writing style of the multiple articles is consistent. The information processing apparatus according to feature 2.

13. The selection unit determines whether the content of the multiple articles will attract the interest of the user viewing the articles. The information processing apparatus according to feature 2.

14. The determination unit determines whether there is a discrepancy in the content of the multiple articles based on whether the content of the multiple articles includes comments classified into the multiple categories. The information processing apparatus according to claim 1 or 2.

15. If the determination unit determines that the content of the multiple articles includes comments classified into the multiple categories, it determines that there is no discrepancy in the content of the multiple articles. The information processing apparatus according to feature 13.

16. A method of information processing performed by a computer, The process of obtaining multiple comments on a trending news article, A classification step for classifying the multiple comments obtained, A first generation step of generating multiple articles from the classified multiple comments, A determination step of comparing the classification results of the classified comments with the content of the articles to determine whether or not there is a discrepancy in the content of the articles, If the aforementioned discrepancies occur in the content of the multiple articles, a second generation step is performed to regenerate the articles in which the discrepancies occur, An information processing method characterized by including

17. Procedure for obtaining multiple comments on a trending news article, A classification procedure for classifying the multiple comments obtained, A first generation procedure for generating multiple articles from the classified multiple comments, A determination procedure for determining whether there is a discrepancy between the content of the articles and the classification results of the classified comments, If the aforementioned discrepancies occur in the content of the multiple articles, a second generation procedure is performed to regenerate the articles in which the discrepancies occur, An information processing program characterized by causing a computer to execute it.

Images