JP2024015904A - News distribution system and news distribution method - Google Patents

Abstract

To provide a news distribution system and a news distribution method that ensure transparency of a sender's position and distribute accurate and fair news.SOLUTION: The method is configured to: collect related articles including primary sources before media edits are made on a news article from a network; determine the accuracy of the news article based on the determined news article and the collected related articles; calculate a digitized fairness score of the fairness of the determined news article and the accuracy of the determined news article; maintain a fairness management ledger associating the fairness score with the news article, and a fairness contribution management ledger associating a user with a fairness contribution degree, i.e, a fairness score assigned to the user when the user performs the behavior contribute to increase or decrease of the fairness score; and reward a user who performs the behavior to increase the fairness of the news article using the fairness management ledger and the fairness contribution management ledger.SELECTED DRAWING: Figure 1

Description

The present invention relates to a news distribution system and a news distribution method.

Fake news, intentional or unintentional misinformation, malicious and inaccurate information, and misinformation have become social problems. With the spread of smartphones and SNS, this has become a familiar issue. Newspaper articles also explain that excessive government guidance on speech risks leading to a controlled society, and that transparency by businesses is important in distributing fair and accurate information. The government also states that ``intervention should be extremely cautious in the interest of freedom of expression,'' and the principle is for the private sector to resolve issues autonomously.

In general, media businesses rely on advertising fees from sponsors, and it is important to write articles that increase the number of readers or page views (PV). This is the so-called attention economy. News articles are basically consumed instantaneously, and amidst a flood of diverse information, breaking news and interesting information are handled in an uncertain manner in order to grab the attention of users, that is, increase the number of PVs. There is a risk. For example, there are cases where information posted on SNS is re-posted without verifying its authenticity.

From the perspective of social responsibility, the media is making efforts to write fair and accurate articles in order to prevent the spread of rumors, and is considering ways to provide technical support. For example, Patent Document 1 discloses a technology that extracts feature amounts from articles, takes into account trends and number of views, and selects whether information should be published. Patent Document 2 proposes a technology that improves the accuracy of articles through a system that awards and deprives users of articles based on authenticity determination, and is a method for mutually verifying authenticity among SNS users. Techniques have also been proposed for verifying the authenticity of information over a network between users.

Japanese Patent Application Publication No. 2020-024489 Japanese Patent Application Publication No. 2019-185768

The spread of SNS has made it possible for information providers to provide information directly to users. Existing media have lost their superiority in terms of breaking news and conveying information as it is, but they are expected to play a role as a facilitator in gathering and organizing information and assembling diverse opinions. However, in news formats where the number of page views is important, it is difficult to incentivize people to spend a long time and carefully write fair and accurate news articles. Here, fairness is slightly different from accuracy and factuality, and includes a subjective concept. Even if there is only one fact, there is a large amount of information that is ``correct for you'' because backgrounds and ways of thinking differ depending on the means of communication and the recipient. It has been pointed out that users are at risk of being exposed to fake news while searching for accurate information. In particular, it is easy to assume that an article is true if it is found to be valid based on one's own information literacy. In other words, it is necessary to create a news distribution system that eliminates obvious distortions of facts and ensures accuracy and fairness, and to achieve a high level of fairness, transparency of processing including the process of selecting information, In other words, it is necessary to ensure transparency in clarifying the perspective on which one's opinion is based.

An object of the present invention is to provide a technology that can ensure transparency of a sender's position and deliver accurate and highly fair news.

The news distribution system according to the present invention is a news distribution system in which a computer executes a program to distribute predetermined news, and monitors news sites on a network and responds to news articles published on the news sites. a buzz news detection unit that detects buzz news whose quotation or comment satisfies predetermined conditions; and a transaction including comments on the buzz news written on the network in association with the buzz news and cumulative values of the comments. A sensitivity information collection unit that collects data, analyzes the comments included in the transaction data, and if the number of clusters obtained by clustering the opinions included in the comments satisfies a predetermined condition, the comment a sentiment analysis unit that determines that news articles with a ``mark'' are highly fair news articles; and a related article that collects related articles from the network, including primary sources for the news article before it is edited by the media. an accuracy determination unit that determines the accuracy of the news article based on the news article determined by the collection unit, the sensitivity analysis unit, and the related articles collected by the related article collection unit; a fairness determination unit that calculates a fairness score that quantifies the fairness of the news article determined by the sensitivity analysis unit and the accuracy of the news article determined by the accuracy determination unit; A fairness management ledger that associates the news article with the user, and a fairness contribution degree that is a fairness score that is given to the user when the user performs an action that contributes to an increase or decrease in the fairness score. a fairness score management unit that manages the associated fairness contribution management ledger, and an action that increases the fairness of the news article using the fairness management ledger and the fairness contribution management ledger. a fairness reward distribution unit that rewards users who have
It is configured as a news distribution system characterized by having the following.

According to the present invention, it is possible to ensure the transparency of the sender's position and to distribute accurate and highly fair news.

Overall processing Processing flow Functional block diagram interface screen Accuracy judgment section Fact checking department Fairness Judgment Department Fairness management ledger for each news article expert database Management ledger of fairness contribution for each user data block news database Hardware configuration Processing of the accuracy determination unit when receiving information from a user

Examples will be described below with reference to the drawings. The embodiments described below do not limit the claimed invention, and all of the elements and combinations thereof described in the embodiments are essential to the solution of the invention. is not limited.

That is, the following description and drawings are illustrative for explaining the present invention, and are omitted and simplified as appropriate to clarify the explanation. The present invention can also be implemented in various other forms. Unless specifically limited, each component may be singular or plural.

The position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. in order to facilitate understanding of the invention. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

In the following description, various information may be described using expressions such as "database," "table," and "list," but various information may be expressed using data structures other than these. "XX table", "XX list", etc. are sometimes referred to as "XX information" to indicate that they do not depend on the data structure. When describing identification information, when expressions such as "identification information", "identifier", "name", "ID", and "number" are used, these expressions can be replaced with each other.

When there are multiple components having the same or similar functions, the same reference numerals may be given different suffixes for explanation. However, if there is no need to distinguish between these multiple components, the subscripts may be omitted in the description.

In addition, in the following explanation, processing performed by executing a program may be explained, but the program is executed by a processor (for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit)), The processor may be the main body of the processing in order to perform the processing using appropriate storage resources (for example, memory) and/or interface devices (for example, communication ports). Similarly, the subject of processing performed by executing a program may be a controller, device, system, computer, or node having a processor. The main body of processing performed by executing a program may be an arithmetic unit or an arithmetic unit, and may include a dedicated circuit (for example, FPGA (Field-Programmable Gate Array) or ASIC (Application Specific Integrated Circuit)) that performs specific processing. You can stay there.

A program may be installed on a device, such as a computer, from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server includes a processor and a storage resource for storing the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. Furthermore, in the following description, two or more programs may be realized as one program, or one program may be realized as two or more programs.

In the first embodiment, a configuration will be described that implements the flow until highly fair news is distributed as described in the problem solving method.

Figure 1 shows the overall flow of the fair news distribution system. The fair news distribution system 1000 (FIG. 3) detects news that has attracted a lot of attention and received a lot of feedback (hereinafter referred to as buzz news) (S101). Next, when the fair news distribution system 1000 detects buzz news, it collects public reactions to the news articles included in the buzz news, analyzes comments and reactions posted by users, and compiles them as statistical data. A sensitivity analysis is performed to visualize various opinions (S102).

Subsequently, the fair news distribution system 1000 evaluates the accuracy and fairness of the news article detected in S101 (S103). Assessing the fairness of a news article means evaluating whether it incorporates information from a variety of viewpoints. In order to perform a fairness evaluation, the fair news distribution system 1000 preferably detects that the content of a news article includes claims from multiple viewpoints and comprehensively analyzes the content. To evaluate the accuracy of a news article, the original source that is the basis for the news article, such as a written document or audio recording of the entire content spoken by the sender, is searched from various sites on the network. As a result, the numbers and original sources are compared with the news article being evaluated to check whether the intent has been altered. Furthermore, the fair news distribution system 1000 evaluates whether the information on which the news article is based has been disclosed, and if so, the reliability of the information. One criterion for evaluating reliability is, for example, whether the information on which the news article is based has been peer-reviewed in order to be published in a public forum such as a journal or a prestigious academic conference. In this case, in the case of academic conferences, the higher the impact factor, the more reliable the information is, and in the case of papers, the number of citations can be used as an indicator. The fair news distribution system 1000 may automatically detect these through a program, but in order to contact an industry expert, the fair news distribution system 1000 may transmit the information on which the above news article is based, and the expert may perform a fact check. Then, it is also possible to receive evaluations and comments from the above-mentioned experts when fact-checking the information.

A comprehensive numerical value of both fairness and accuracy evaluation results is called a fairness score. If the discourse only takes up some of the diverse opinions and is not comprehensive, the fair news distribution system 1000 determines that it does not meet a certain standard and calculates a low fairness score. On the other hand, when the information on which the judgment is based is clear and the factuality is high to a certain degree, the fairness news distribution system 1000 has a high accuracy evaluation, and therefore calculates a high fairness score as meeting a certain standard. . In other words, even if a news article is biased, it can be made transparent by covering multiple points of view and making it clear which of those points of view your opinion is based on. The more factual the article, which is logically explained based on the information on which the news article is based, the higher the fairness score will be.

After evaluating the fairness and accuracy of the news article in S103, the fair news distribution system 1000 formats the information included in the news article to generate a site that serves as a reference information source for the user to browse, and then displays the news article. (S104). It is assumed that reference information sources will be updated sequentially. For example, immediately after buzz news is detected, it is assumed that experts have not finished fact-checking and the determination of accuracy is incomplete. Even in this case, the fair news distribution system 1000 describes the progress of the fact check as "determining" or the like, and creates a thread of the buzz news being verified on the reference information source site. The thread includes objects that display the results of the sensitivity analysis performed in S102, the accuracy and fairness evaluation results performed in S103, and the like. Each object will be described later using FIG. 4.

As described above, in the process of generating a reference information source, the fair news distribution system 1000 scores fairness contribution degrees (described later) for users who have contributed to generating fair and accurate information. Then, the fair news distribution system 1000 gives a reward according to the score to realize an incentive for the framework that provides fair information (S105). The fair news distribution system 1000 continues the processing of S102 to S105 until the fairness score becomes equal to or greater than a predetermined threshold or a predetermined time elapses (S106). If the fair news distribution system 1000 determines that the fairness score has not exceeded the predetermined threshold or that the predetermined time has not elapsed (S106; No), the process returns to S102 and continues the subsequent processing. On the other hand, the fair news distribution system 1000 ends the process when it is determined that the fairness score has exceeded a predetermined threshold or that a predetermined time has elapsed.

The fair news distribution system 1000 may track people's reactions more closely by shortening the interval at which the emotional analysis information 502 is checked in S102 as the buzz news attracts more attention. Alternatively, in S103, the fair news distribution system 1000 may increase the number of people who request fact checking or perform control to encourage the provision of information on verification data in order to improve accuracy judgment accuracy.

A sequence for realizing the above processing and a functional block diagram for realizing it will be explained using FIGS. 2 and 3. FIG. 2 is a sequence diagram specifically showing the flow of the process shown in FIG. FIG. 3 is a block diagram showing the functional configuration of the fair news distribution system 1000. As shown in FIG. 3, the fair news distribution system 1000 includes a back end section 1001, a front end section 1002, and a fairness management section 1003.

These may be operated in unison, but in order to ensure transparency, it is better to have a structure that allows disclosure of the parts related to increases and decreases in fairness and the distribution of rewards for actions that contribute to fairness. For example, the front end section 1002 and the back end section 1001 may be separate nodes on the same platform, and the fairness management section 1003 may be operated as a separate node in order to entrust its operation to a third party organization. Communication between each node is preferably configured to allow disclosure. This is considered to be compatible with transactions using blockchain B's smart contracts and data management using ledgers.

As shown in FIG. 3, this system mainly includes a buzz news detection unit 201 having a function of analyzing the original source that is the basis of the buzz news article and tracing the origin of the original source; A fact checking unit 207 that requests fact checking from experts to confirm that it does not contain false information; a sensitivity analysis unit 203 that has a function of evaluating the bias of the original source of the buzz news; A fairness determination unit 206 has a function of evaluating the fairness of news articles and reporters who have created them, and outputs the output results obtained from the processing results of each of these units as a viewable screen or website along with past history. It is characterized by having a display information generation unit 209, a fairness reward distribution unit 210 that manages fairness evaluation results in a blockchain B ledger and distributes tokens as rewards based on fairness. Below, there are two types of fairness scores: fairness scores given to news articles and fairness scores given to people. A person is given a fairness score when they perform an action that contributes to an increase or decrease in the fairness score given to a news article. Therefore, the fairness score given to a person is called the fairness contribution degree.

This system can be realized using general information processing devices such as servers and terminals as hardware. An example of the hardware configuration will be described later using FIG. 11.

Further, various data stored in the information processing device or used for processing can be realized by the CPU reading out and using the data from the memory. In addition, each functional unit of the information processing device (for example, the buzz news detection unit 201, fact check unit 207, sentiment analysis unit 203, fairness determination unit 206, fairness reward distribution unit 210, etc.), each function shown in FIG. Part) can be realized by the CPU loading a predetermined program stored in an external storage device into memory and executing it.

In the following, a case will be exemplified in which this system is configured by one computer, but all or part of these functions are distributed over one or more computers such as a cloud, and are provided via a network. Similar functions may be realized by communicating with each other.

The blocks that implement the above functions will be explained in detail below. Here, assume that the media server 1200 distributes news articles via the Internet, which has become common in recent years, and that the distributed news articles have attracted a large amount of user interest at a certain portal site 1300. In fact, existing news media also have a structure in which experts or users add comments to news articles of high interest (S300).

The buzz news detection unit 201 of this system detects the news that attracted a large amount of interest in S300 as buzz news (S301). The buzz news detection unit 201 monitors news sites and SNS on which news articles are posted, and detects whether the number of citations and the number of comments of a news article are greater than a predetermined threshold. Of course, if a news article that meets predetermined conditions is detected, it may be assumed that buzz news has occurred. Hereinafter, the term "news site" includes various network services such as SNS where news articles are published.

Buzz news indicates a high degree of public interest, and it is assumed that there is a high demand for accurate and fair information. There are two actions you can take to capture unbiased information. This involves evaluating the accuracy and fairness of the news articles themselves, collecting public opinions, reflecting diverse opinions, and redistributing information after supplementing missing viewpoints.

Therefore, the emotional information collection unit 202 collects the user's reactions (S302). The emotional information collection unit 202 collects transaction data including comments on news sites, portal sites, SNS, cumulative values of reaction buttons, etc. as user reactions. The viewing mode of the site includes not only an Internet browser but also an app. Buzz News often has a large number of clicks on the "like" button, retweet, and other quote buttons. Impressions are often expressed following the quotation, and the sensitivity information collection unit 202 can efficiently collect comments from users associated with the buzz news by following the link to the buzz news. Conceivable.

The sensitivity analysis unit 203 is in charge of analyzing, visualizing, and quantifying the user's reaction information collected as described above (S303). The evaluation axis for visualization depends on the topic of the news article, but for example, in the case of political topics, attributes include right-wing/left-wing, hawkish/dovish, and support/opposition to a policy. For other evaluation axes, for example, sensibility analysis may be used, with emotional analysis, moral analysis, surprise analysis, etc. as evaluation axes. The sensitivity analysis unit 203 preferably analyzes and visualizes the user's reaction from various viewpoints as described above. The sensibility analysis unit 203 should use machine learning to classify these, but for example, after extracting keywords from news articles, name matching and removing noise from keywords that have different meanings depending on the context may be used. This technology is required.

The sensitivity analysis unit 203 preferably collects user reactions to the extent that it is possible to visualize that a plurality of opinion clusters are formed and there are various opinions. The sensitivity analysis unit 203 clusters the opinions of various users, analyzes and visualizes which opinion cluster the Buzz News reporter belongs to and which viewpoint the article is from, thereby detecting bias in the news. It is possible to grasp it while being aware of gender. This is considered to be a state of high fairness. For example, the sensitivity analysis unit 203 clusters various opinions of users regarding a news article, and if the number of clusters obtained by the clustering is equal to or greater than a predetermined threshold, the sensitivity analysis unit 203 determines that the news article is a news article with high fairness. I judge that there is. For the above-mentioned clustering, various conventionally known techniques may be used, such as the Ward method, the group average method, the shortest distance method, and the longest distance method.

On the other hand, the related article collection unit 204 collects related articles related to the news article in addition to the original news article included in the buzz news detected in S301 (S304). Here, the related article refers to, for example, the original source used for the news subject, past similar news and analysis articles, and literature/distribution that serves as the basis for the news article. Distribution, which serves as the basis for a news article, refers to raw data before media editing, including primary sources published by government or corporate public relations, for example. The related article collection unit 204 may also target audio data and video data, and may also capture information other than text media by using technology such as character transcription. Text transcription technology has made great progress in recent years, and it is believed that audio and video data generated in text can be put to practical use by appropriately passing it through a noise filter.

These items may be collected manually by the operator of this system, may be collected according to a predetermined algorithm, or may be collected based on posts from users. As for the method of relying on posts from users, for example, the fairness information UI (User Interface) 208 accepts posts from users and sends the contents to this system, or the sensitivity analysis unit 203 analyzes posted comments. It may be extracted during the process. The sensibility analysis unit 203 collects information on these related articles into the article DB 3001. Then, the accuracy determination unit 205 analyzes the related articles, and the display information generation unit 209 outputs the results in a list for display on the fairness information UI 208 screen.

The accuracy determination unit 205 diagnoses the accuracy of the news article based on the news article for which reaction information was collected in S302 and the related articles collected in S304 (S305; accuracy evaluation). At the time of diagnosis, the accuracy determination unit 205 performs tagging, gist extraction, etc. on the news articles and documents imported by the related article collection unit 204 so that they can be searched as past similar news, and It is a good idea to import and manage it. The accuracy determination unit 205 evaluates the accuracy of the news article by performing processing in the flow shown in FIG. In this process, the accuracy determination unit 205 diagnoses whether the gist of the news article differs from the related article, which is the collected primary source, and whether the news article is written logically. The purpose is to prevent news articles that manipulate impressions by cutting out only part of the primary source.

FIG. 5 is a flowchart illustrating an example of processing related to accuracy evaluation in S305 illustrated in FIG. As shown in FIG. 5, the accuracy determination unit 205 analyzes the news article with reference to the article DB 3001 (S501), and finds that there is a description disclosing the primary source in the news article, and that the primary source is actually the primary source. It is determined whether it is possible to access and view (S502). If it is determined that the above-mentioned viewing is possible (S502; Yes), the fairness score of the news article is added (S503), and the primary source is analyzed (S504).

On the other hand, when the accuracy determination unit 205 determines that the above-mentioned browsing is not possible (S502; No), the accuracy determination unit 205 determines whether the related article collection unit 204 is able to search the primary source (S505). If the accuracy determining unit 205 determines that the related article collecting unit 204 has been able to search the primary source (S505; Yes), the accuracy determining unit 205 analyzes the primary source searched by the related article collecting unit 204 in S504. When analyzing a primary source, the accuracy determining unit 205 detects omissions in the main point, different interpretations, and simple mistakes by reading each of the primary source and the news article, and comparing and analyzing the two. These processes may be performed algorithmically or by the operator of this system. To ensure transparency, the accuracy determination unit 205 leaves the process of how the determination was made (for example, the order and method of the comparison, and the points where the two differ). When using machine learning, it is best to use models that can be explained, such as by displaying the importance of features.

As a result of the analysis in S504, the accuracy determination unit 205 determines whether the content of the primary source has been altered (S506). If the accuracy determination unit 205 determines that the content of the primary source has not been altered (S506; No), it adds a fairness score (S507). On the other hand, if the accuracy determination unit 205 determines that the content of the primary source has been altered (S506; Yes), it significantly lowers the fairness score by a value that is greater than or equal to a predetermined threshold and considers the report to be malicious. (S508).

The accuracy determination unit 205 outputs a fairness score regarding accuracy that reflects the values calculated in S507 and S508 (S509). Further, a fairness contribution level corresponding to the fairness score is output to the reporter of the analyzed news article (S510).

Returning to FIG. 2, the accuracy determining unit 205 further outputs the primary source and the news article to the fact checking unit 207, and the fact checking unit 207 diagnoses the accuracy of the news article. In diagnosing accuracy, it is important to incorporate the opinions of external experts. Therefore, the fact check unit 207 transmits the primary source and the news article to the expert's site or terminal, and issues a fact check request (S305; fact check request). The fact check unit 207 uses a pre-prepared expert DB (database) 3002 as the criteria for selecting a party to whom it should send a fact check request, based on a specialist in an occupation that matches the field of the news article. You may select your own home, or you may select from experts who have contributed to improving the fairness of similar news in the past based on a ledger that is distributed and managed on the blockchain.

FIG. 8B is a diagram showing an example of an expert DB. As shown in FIG. 8B, the expert DB 3002 contains the expert ID for identifying the expert, the expert's name, the expert's contact email address, the expert's occupation, and past fact information. It is associated with past results indicating that the check has been performed. In FIG. 8B, for example, the expert "T.Y" identified by the expert ID "S0000001" is a university professor, the email address registered as a contact is "α", and he has previously fact-checked This indicates that the company has a track record of receiving requests from customers.

Experts who have received a fact check request from this system will respond with supporting information and opinions by operating computers such as fact-checking terminals and servers, and will submit the fact-checking results to this system. Send (S307). From the viewpoint of fairness, diverse opinions are necessary, so in S305, the fact check unit 207 sends the primary source and news article to the expert's site or terminal, and requests opinions from multiple experts. It is better to make it . The information presented by the expert and its source can also be used as a data collection source for the related information collection unit 204, and the fact check unit 207 may preferably register it in the article DB 3001 for the next time onwards.

FIG. 6 is a flowchart illustrating an example of the process related to the fact check request in S305 illustrated in FIG. As shown in FIG. 6, the fact check unit 207 analyzes the primary source to be transmitted and the genre of the news article (S601), and selects an expert who specializes in the genre obtained through the analysis from the expert DB 3002. (S602). The fact check unit 207 sends the primary source and news article scheduled to be sent to the email address or site associated with the selected expert, and requests a fact check (S603).

The fact check unit 207 receives the fact check result from the expert who requested the fact check, and determines whether the content has been evaluated as good or appropriate (S604). For example, if the above fact check results include negative comments such as the fact that the news article is likely to be false, it is not possible to judge whether the content was evaluated as good or appropriate. If a positive comment such as an evaluation or an evaluation that the news article is not valid, or a positive comment such as that the news article is likely to be true, may be evaluated as a good evaluation or an evaluation that the news article is appropriate.

If the fact check unit 207 determines that the content has been evaluated as good or appropriate (S604; Yes), the fact check unit 207 determines that the news article is accurate and sets the fairness score of the news article by a predetermined value. is added (S605), and the process proceeds to S606. On the other hand, if the fact check unit 207 determines that the content is not rated well (S604; No), the fact check unit 207 determines that the news article is not accurate and proceeds to S606 without doing anything.

When the predetermined value is added to the fairness score of the news article, the fact check unit 207 updates the fairness score (accuracy) of the article to the fairness score to which the predetermined value is added (S606), and A predetermined value is added to the fairness contribution degree of the expert who performed the check, and the value is updated (S607). Through the above processing, it is possible to calculate accuracy and fairness contribution among the fairness score indicators.

Returning to FIG. 2, the fairness determination unit 206 comprehensively digitizes both the fairness and accuracy evaluation results to calculate a fairness score (S308). The fairness determining unit 206 calculates a low fairness score if the discourse only picks up some of the diverse opinions and is not comprehensive. On the other hand, if the information on which the judgment is based is clear and highly factual, the accuracy will be evaluated high, so a high fairness score will be given. In other words, even if an article is considered to be so-called biased, it must be transparent enough to cover multiple viewpoints, clarify which viewpoint one's own opinion is based on, and have clear argument data. If the thesis is used correctly, the article will be considered logical, that is, highly factual, and will be judged as an article with a high fairness score.

For example, in S308, the fairness determination unit 206 may calculate the fairness score using a flow as shown in FIG. The fairness determination unit 206 obtains various opinion clusters analyzed and acquired by the sensitivity analysis unit 203 (S701). The fairness determination unit 206 analyzes the intention of each opinion cluster, classifies the viewpoints, and attaches identifiers of users belonging to each opinion cluster. For example, the fairness determination unit 206 associates multiple user identifiers (user IDs 001, 002, . . . ) with one opinion cluster A. Furthermore, the fairness determining unit 206 associates the identifiers of a plurality of users (user IDs 001, 003, . . . ) with another opinion cluster B. Then, the fairness determination unit 206 analyzes the news article of Buzz News, and calculates the degree of similarity between the two to determine which opinion cluster the news article was written from a viewpoint close to that of the other (S703).

Further, the fairness determining unit 206 determines whether the Buzz News news article refers to opinions belonging to a plurality of opinion clusters and then draws a conclusion. That is, the fairness determining unit 206 determines whether the news article is analyzed from various viewpoints (S704).

If the fairness determining unit 206 determines that the news article is analyzed from various viewpoints (S704; Yes), it determines that the news article is unbiased and adds a fairness score (S705). . Bias in this system means that news articles are summarized from a single point of view. The fairness determining unit 206 considers the news article to be unbiased and fair if it reaches a conclusion after touching on arguments from multiple viewpoints. For example, the fairness determination unit 206 determines the condition that a certain news article is an article that describes the opinions of more than half of the opinion clusters among the plurality of opinion clusters, and also states one's own opinion. If the conditions are satisfied, the news article is determined to be an unbiased and fair news article. Furthermore, when the above conditions are met and the fairness score is added, the fairness determination unit 206 adds a higher fairness score as the number of opinion clusters increases.

Overall, a fair news article in this system is one that does not include unnecessary changes from the primary source, and that reduces bias by touching on the arguments of multiple viewpoints before reaching a conclusion. This refers to articles that meet two requirements. A fairness score, that is, a degree of fairness contribution, is given to a person who has taken an action to compensate for the lack of these requirements (or an action that increases these requirements to a certain degree).

As explained above, in this system, the fairness of an article is clarified in the process of calculating the fairness score. At this time, in order to take into account the degree of contribution to fairness, this system also provides tokens as rewards based on the degree of contribution to fairness. This provides an incentive for the system operator to operate the system in the public interest.

The diversity of viewpoints included in the opinion cluster obtained by the sensitivity analysis in S701 has a high degree of contribution to fairness. Therefore, there may be a mechanism for redistributing tokens acquired by the system to users who have provided information constituting opinion clusters (312). To make this possible, ID coordination is required between this system, another portal site or SNS server 1300 that can post comments, and the user who operated the user terminal 1400 to specify the user who provided the above information. That is, in S302, the emotional information collection unit 202 collects the user ID of the user who has added a comment (that is, the comment added in S300 shown in FIG. 2) to the buzz news detected in S301, along with the user's reaction. , and register it in this system. By linking user IDs, the user who provided the above information (or the user who intends to provide the above information in the future) can feel free to return their reactions on SNS, and the tokens will be redistributed. , it is possible to realize a system in which participants can be recognized as participating in actions to improve fairness and automatically receive incentives. This can be expected to encourage user participation.

Further, in S304 shown in FIG. 2, the related article collection unit 204 collects related articles in addition to the original news article included in the buzz news, but the media server 1200 collects related articles via the information providing interface 508 in FIG. Alternatively, the corrected article may be published online or offline (S306a). Normally, when the media publishes a corrected article as a follow-up, it not only reduces the space available for transmitting new information, but also requires a great deal of effort. Therefore, in this system, the publication of a corrected article is evaluated as an action to improve fairness, and in S312, tokens serving as rewards are also distributed to the media. This will encourage the media to publish corrected articles and disseminate additional information, making it possible to provide supplementary news distribution that goes beyond the traditional news distribution of the fastest transmission of information.

Fairness scores can be managed using a blockchain ledger to track and provide transparency when changes are made to the system. While creating an environment that is less susceptible to information manipulation, it will be possible to ensure transparency regarding actions that contribute to improved fairness and provide the game experience of collecting tokens at the same time. Controlling the degree of fairness contribution is sensitive and high-risk information that affects the creation of public opinion in society, and robustness is more important than scaling, so a private type is considered to be more advantageous. It is desirable that this information be operated and managed by a non-profit organization.

The fairness reward distribution unit 210 and the fairness score management unit 211 manage the fairness contribution of users and the fairness score of news articles, and control transactions regarding increases and decreases in these on blockchain B. Along with these transactions, the evaluation of actions that contributed to fairness and the amount of updated scores are recorded along with the basis. Data managed by the distributed ledger is shown in FIGS. 8A and 9.

The fairness reward distribution unit 210 and the fairness score management unit 211 assign fairness contribution degrees to experts who cooperated in fact checking. The fairness contribution degree may be issued using, for example, tokens distributed on the blockchain. It is desirable that the history of fairness contributions and token exchanges be managed on the blockchain from the perspective of ensuring transparency and tamper resistance. As described above, the fairness reward distribution unit 210 and the fairness score management unit 211 may be operated by a fairness management unit that is independent of the back-end logic unit. If the provider of the back-end logic section and the provider of the fairness management section are the same, there is no need for division.

The fairness score management unit 211 preferably maintains a fairness management ledger 801 for each news article, as shown in FIG. 8A. The fairness management ledger 801 is a table for managing the fairness of news articles. As shown in FIG. 8A, the fairness management ledger 801 includes the news ID for identifying buzz news, the fairness score of the published news article included in the buzz news, and the fairness of the original news article of the buzz news. The raw score that is the fairness score, the last update date of the fairness score, and the transaction ID for identifying transaction data when reactions to the news identified by the news ID are collected are stored in association with each other. Ru. The above-mentioned raw score is a fairness score before reflecting an increase or decrease in the degree of fairness contribution to the fairness score of the original news article of Buzz News.

In FIG. 8A, for example, the fairness score of the news article published as Buzz News identified by the news ID "N0000001" is calculated as "80", and the raw score of the original news article of the Buzz News is " 55". Further, the date and time when the fairness score "80" was most recently updated was "May 20, 2021", indicating that it was updated using transaction data identified by the transaction ID "T00000001, T00000003". Specifically, as will be described later, by displaying the evaluation of accuracy and fairness (fairness of news articles) on the same fairness information UI 208 as a result of sensitivity analysis, the above fairness score can be viewed from an objective perspective. This can improve the fairness of news articles. Note that the raw score of the fairness score is assumed to be the value initially calculated by the accuracy determination unit 205.

The accuracy determination unit 205 preferably writes the news article fairness management ledger 801 through a predetermined API (application programming interface) configured on the blockchain. As described in S704 and the like, the fairness determining unit 206 calculates the fairness score regarding the news article and updates it through the above-mentioned predetermined API.

When the fact check unit 207 receives the result of the fact check from the expert, it is preferable that the fact check unit 207 performs procedures related to updating the fairness score of the news article through the above-mentioned predetermined API. In the fairness score update procedure for these news articles, information is exchanged using data blocks 9001 exchanged on a blockchain as shown in FIG. 10, for example. Here, updating of the fairness score for a news article has been described, but the fairness contribution degree, which is the fairness score for a person, can be considered in the same way.

FIG. 10 is a diagram illustrating an example of data blocks included in tokens exchanged on the blockchain. As shown in FIG. 10, a data block 9001 is a history ledger of data blocks exchanged for each transaction ID. The data block 9001 is identified by the transaction ID described above, the date on which the block data constituting the transaction data identified by the transaction ID was exchanged on the blockchain, the type of increase/decrease in the fairness score, and the type. the increase/decrease in the fairness score, the evaluation action that contributed to the increase/decrease in the fairness score, the news ID of the news article for which the fairness score was calculated, and the action date and time that is the date and time when the above evaluation action was performed. , is data in which the name of the user who performed the above-mentioned evaluation behavior and the user ID are associated with each other.

In FIG. 10, for example, the block data constituting the transaction data identified by the transaction ID "T00000002" is reported by the user ID "U0000022" (user name: T.Y.) as an expert on May 18, 2021. A fact check was performed on the article "N0000002", and on the same day, a fairness score (here, the sum of the fairness score for the news article and the fairness score, which is the fairness score for the person, of the fairness contribution degree) was calculated. This indicates that the data has a value of "+15". In this way, in this system, the fairness scores are updated in the same format on the blockchain, so data regarding the fairness evaluation of various news articles can be exchanged with the same level of security. Note that the user name may be a nickname.

The fairness score management unit 211 maintains a management ledger that manages the fairness contribution degree for each user as shown in FIG. Update the fairness contribution information managed by each. The fairness contribution management ledger 9101 is a table for managing the fairness contribution, which is a fairness score for a person.

As shown in FIG. 9, the fairness contribution management ledger 9101 includes the above-mentioned user ID and user name, the type indicating the occupation of the user identified by the user ID and user name, and the last update of the fairness contribution. date, the history of the fairness contribution of the user, the amount of tokens the user has acquired (token amount), and the transaction ID of the transaction data that is the basis for acquiring the amount of tokens. are associated and retained.

In FIG. 9, for example, the user with the user name "T.Y" identified by the user ID "U0000022" is an expert who performs fact checking, and the fairness contribution is calculated with May 18, 20201 as the last update date. , it shows that a token amount of "1500" has been calculated. It also shows that in the past, a token amount of "1485" was calculated for the user on May 10, 2021. Specifically, as will be described later, by displaying the evaluation of accuracy and fairness (fairness to the person) on the same fairness information UI 208 as a result of sensitivity analysis, the above fairness score can be viewed from an objective perspective. This can improve the fairness of news articles.

The fairness score management unit 211 obtains information on the fairness score for each user based on the evaluation actions taken by each user to improve fairness and the increase/decrease in fairness with respect to the original news article of the buzz news. Approve updates to the managed ledger. Processing performed by the system is also managed as a fairness score. The fairness score of the system is also improved when the accuracy judgment unit 205 searches for and discovers a primary source, or when the fairness judgment unit 206 is able to add a new viewpoint that was not present in the news article. It is recognized that there was a function to increase the number of employees, and a level of fairness contribution is assigned. The fairness reward distribution unit 210 prepares tokens that are linked to the degree of fairness contribution and circulate on the blockchain, and distributes them as rewards to users participating in the blockchain.

The problem with this system is how to ensure economic rationality for media and user participation. Since this system is a highly public system, it is required to be operated by a non-profit organization and to have a structure that is not susceptible to influence from specific sponsors such as politics or corporations. Therefore, one incentive is to create a game by collecting tokens. As an example, it is preferable to maintain tables such as those shown in FIGS. 8A to 10, and use a mechanism to manage fairness scores and tokens on the blockchain so that they can eventually be exchanged for currencies outside the blockchain. For example, the fairness reward distribution unit 210 may manage the token in a format that allows the amount of increase/decrease in the fairness contribution and exchange with other currencies.

Furthermore, for example, in the GUI shown in FIG. 4, donations obtained through the donation interface 507 or other routes can be pooled and exchanged for tokens collected by each user. This will allow economic activities managed on the blockchain to go back and forth between economic activities in the real world. In that case, it is preferable that the fairness reward distribution unit 210 provides a series of mechanisms related to exchange in cooperation with an external system.

Returning to FIG. 2, the display information generation unit 209 compiles information in the form of a news summary site so that a variety of information can be viewed at a glance as a reference information source (S309). For example, the display information generation unit 209 constructs a site screen having a fairness information UI 208 as shown in FIG. When constructing the screen, the display information generation unit 209 generates a thread that reflects information calculated midway through each unit such as the fairness determination unit 206, the sensitivity analysis unit 203, the accuracy determination unit 205, and the fact check unit 207. The generated thread is stored in the display DB 3003. In FIG. 4, the screen includes a thread 401 that was updated on May 15, 2021, and is stored in the display DB 3003. When constructing a screen, the display information generation unit 209 reads threads stored in the display DB 3003 and generates the screen. For example, when an input of a certain keyword "safety of coronavirus vaccine" is accepted, threads about buzz news articles that are somewhat similar to the keyword are extracted from the display DB 3003 and displayed on one screen. The above screen may be provided by registering the output results of each section such as the news article, the sentiment analysis section, the information verification section, and the fairness judgment section, along with past history, on a viewable website. The above screen will be specifically explained below.

FIG. 4 is a screen showing an example of the fairness information UI. As shown in FIG. 4, the fairness information UI 208 preferably displays a Verification object 501 with a link so that the user can refer to the primary source information collected by the sentiment analysis unit 203 and past similar news. The Verification object 501 may also display information on the fairness contribution calculated by the accuracy determination unit 205. In FIG. 4, the display information generation unit 209 determines that two documents, reference [1] ○○ and reference [2] △△, are the primary sources for the Buzz News news article with the title "Safety of Corona Vaccine". This indicates that the information is displayed on the fairness information UI 208 as information collected as follows. It also shows that the fairness contribution to the news article was calculated as "+5".

In addition, the accuracy determination unit 205 analyzes the accuracy of the original news article of the Buzz News news article. Therefore, the display information generation unit 209 displays the result on the fairness information UI 208 in parallel with the text of the original news article of Buzz News, like the object 503, as the raw score before improving the fairness. good. In FIG. 4, it can be seen that the display information generation unit 209 displays "65" as the raw score in association with the text of the original news article of the buzz news written by reporter A.

Further, the display information generation unit 209 may cause the fairness information UI 208 to display an object 504 that displays the judgment status of the fact check by the expert. The display information generation unit 209 reads the fact check results output from the fact check unit 207 and classifies them into three states: under determination, determined, and undetermined. When the determination has been completed, the display information generation unit 209 adds information on the degree of fairness contribution and updates the display, assuming that the fairness of the display content of the page has increased. In FIG. 4, it can be seen that the status is "determining", which indicates that the fact check is being executed.

Further, the display information generation unit 209 may display an object 502 that illustrates an opinion cluster, a reaction summary, or an emotion such as anger or joy using a meter as the output result of the sensitivity analysis unit 203. In FIG. 4, the emotion meter outputs the fact that among the opinion clusters obtained through processing by the sensitivity analysis unit 203, there are more opinion clusters indicating "joy" than there are opinion clusters indicating "anger." Which emotion an opinion cluster represents can be realized, for example, by semantically analyzing and aggregating the viewpoints of each opinion cluster classified by the fairness determining unit 206. As a result of the sensitivity analysis provided by this system, if an opinion cluster is formed from a viewpoint that was not included in the article, it is preferable to display the degree of fairness contribution of the users who sent the opinions forming the opinion cluster.

Furthermore, in the fairness score transition object 505, the display information generation unit 209 calculates the fairness contribution degree given by each functional block using the raw score information (FIG. 8A) of the original news article of the buzz news as a starting point. It would be good to have a mechanism that can display in chronological order how the fairness of the displayed content of the page has been updated as the numbers are added up. For example, it is preferable to display the fairness contribution added by analyzing the original news article of Buzz News and the overall fairness score of the information currently displayed on the page. The overall fairness score is the score obtained by adding the added fairness contribution (+15) to the starting point fairness score, and is displayed as "80" in this example. In this way, the raw score of the original news article of Buzz News and the fairness score after increasing or decreasing the fairness contribution are displayed as the current overall fairness score, and as shown below. The changes in the increase and decrease mentioned above are output on the screen. Here, in order to ensure transparency of what actions contributed to fairness when there is a change in the time-series value of fairness, we have designed a GUI that can trace logs. In FIG. 4, a graph showing a temporal transition from the starting point fairness score S1 to the current comprehensive fairness score S2 is displayed.

Further, the display information generation unit 209 preferably displays on the object 506 the behavior history of all users who contributed to the change in the fairness score for each Buzz News news article. Information related to increases and decreases in the fairness score may be managed on the blockchain (for example, FIGS. 9A and 9B), and the display information generation unit 209 may refer to it and retrieve it when creating the object 506. In this example, the display information generation unit 209 reads the fairness contribution degree management ledger 9001 and displays in the object 506 the details of the users who contributed to the increase/decrease in the fairness contribution degree.

The display information generation unit 209 uses the transaction ID listed in the fairness management ledger 801 for each news article as shown in FIG. 8A as a key as information to be displayed on the object 506, and evaluates which behavior of which user. and show whether the fairness has been increased or decreased.

The fairness information UI 208 of this system is a mechanism that accumulates threads consisting of news articles of each buzz news and their analysis results (corresponding to objects 501 to 506) as time passes, but in thread units. It is preferable to provide an interface 507 that allows donations to be made. In FIG. 4, for example, when the display information generation unit 209 receives an operation (for example, a click) on the interface 507, the display information generation unit 209 calculates the amount corresponding to the operation (for example, the amount obtained by multiplying the number of clicks by assuming that one click is 100 yen). and outputs it as a donation amount (S311). The display information generation unit 209 converts the donation amount into a fairness score at a predetermined ratio. There are several possible distribution methods after this. Since donations themselves do not directly contribute to improving the fairness of news articles, the amount of tokens held by users is managed as shown in Figure 9, and donations can be made in bulk when a user requests an exchange. Another option is to exchange tokens. This role is played by the fairness reward distribution unit 210. On the other hand, there is also a method of recording the above-mentioned fairness increase/decrease in the transaction data and distributing it to the information providers. By managing the fairness score as a token distributed on the blockchain, it will be possible to provide a reward distribution function that is resistant to tampering. In FIG. 10, the user "S.N" identified by the user ID "U0000003" views the news article of Buzz News identified by the news ID "N0000003" on May 22, 2021, and changes the fairness. The donation amount equivalent to the amount "+10" is donated, and the display information generation unit 209 outputs it as transaction data identified by the transaction ID "T0000001".

The display information generation unit 209 preferably attaches an interface 508 that provides information that contributes to improving fairness to each article thread, similar to the donation interface 507. The use of the fair information provision interface 508 is to provide information on primary sources to be referred to by the accuracy determination unit 205, collect user opinions that serve as material for the sensitivity analysis unit 203, and receive information from experts in the fact check unit 207. etc. In FIG. 4, for example, when the display information generation unit 209 receives an operation on the interface 508 (for example, an operation to attach and send a primary source to be provided by clicking on the information providing icon), the display information generation unit 209 receives an amount ( For example, assuming that one document is 1,000 yen approved by the government, an amount corresponding to the number of documents is calculated and output as the information provision amount. As in the case of the interface 507, the display information generation unit 209 converts the information provision amount into a fairness score at a predetermined ratio, records it in the fairness increase/decrease of the transaction data described above, and distributes it to the information provider. do. In this case, as in the case of the interface 507, the fairness score is managed as a token distributed on the blockchain, making it possible to provide a reward distribution function that is resistant to tampering. In this case, the fairness reward distribution unit 210 gives a reward based on a fairness score that reflects the increase or decrease in fairness for each of the tokens identified by the transaction ID.

Furthermore, for each article thread, it is preferable to provide a history information call backlog interface 509 that allows the user to view the backlog when he or she wants to refer to past information beyond the displayed information. By selecting the backlog interface 509, the user can refer to display information of threads at a certain point in the past. For example, when the display information generation unit 209 accepts an operation on the backlog interface 509 (for example, clicking a backlog icon and inputting a date to specify a thread to be referenced), the display information generation unit 209 extracts the thread including the date from the display DB 3003. However, it may also be displayed in chronological order on the screen.If a certain keyword (for example, "safety of coronavirus vaccine") is entered on the screen, a thread containing the above date that includes the keyword will be displayed. displayed on the screen.

The display information generation unit 209 reads related news IDs, tag information obtained in the process of analyzing news articles and comments from the article DB 3001 as shown in FIG. 11, and extracts, for example, news articles containing the above keywords. The article DB 3001 is accumulated by the sensitivity analysis unit 203.

FIG. 11 is a diagram showing an example of the article DB 3001. As shown in FIG. 11, the article DB 3001 includes a news ID for identifying a news article of Buzz News, a headline of the news article identified by the news ID, and a related article for identifying an article related to the news article. It includes the ID, tag information attached to the news article, the date and time the news article was published, the date when the news article became a buzz, the name of the reporter who reported the news article, and the organization to which the reporter with the name belongs. For example, in FIG. 11, the headline of the news article identified by the news ID "N0000001" is "After vaccination...", and the related article identified by the related article collection unit 204 by the related article ID "RN000002". indicates that it is the news article that was searched. In addition, the tag information attached to the news article is "vaccine, politics, medical care, corona", and the news article was posted on May 14, 2021 by D. This article was created by U, and indicates that the next day, May 15, 2021, was the day the buzz occurred. Furthermore, the site of the original news article of the buzz news is "http://www.xxx.co.jp/yyy.html", and the references collected as the primary source are reference [1]○○ , Reference [2]△△, indicating that it is published at “http://www.zzz.org/20210514.html”.

In this way, the data that serves as the basis for calculating the fairness score calculated by this system is displayed on the fairness information UI 208. The fairness information UI 208 may be provided as an Internet site or a mobile terminal application (S310).

Just as you can see on social networking sites and news comment sites, when you see information that you can subjectively judge to be convenient or correct, you tend to strongly believe that the information is accurate, and you disagree. They tend to have a strong reaction to people who hold opinions. Fair information is difficult to provide in the form of news distribution, so one of the requirements is that it is organized in a format that allows for comparisons with different opinions. Therefore, like this system, the accuracy of the news article is diagnosed, feedback such as evaluations and comments are obtained from fact-checking of the news article by reliable experts, and opinions from the public obtained through opinion clusters, etc. It is desirable to be able to confirm opinions and feedback on one screen.

Sites that are widely used as word reference information sources are based on the philosophy that "everyone's opinion is surprisingly correct," and that information can be considered "surprisingly correct" if it has aggregation, diversity, independence, and decentralization. We utilize this information to disclose information and information sources from diverse perspectives. The reference information source for the news version provided by this system is widely used from the viewpoints of relying on users to collect evidence, displaying information such as reference information (verification), keeping a history of revisions, and being operated by donations. Close to the site being used. From a different perspective, there is a function to analyze and display user reactions in near-real time, a function to track public opinion that changes over time, a function to display news fairness scores, and a function to display what actions have improved the fairness contribution. One point is visualization. In this system, for example, as shown in the screen shown in Figure 4, public opinion data (e.g., opinion clusters) is corrected and updated in chronological order, ensuring transparency of the sender's position, and ensuring accuracy and fairness. It is beneficial for users that high-quality news is distributed.

The degree to which you can obtain fair information by reading this article comprehensively is called the overall fairness score. The display information generation unit 209 displays the fairness score of the original news article of Buzz News and the fairness improved in the process of putting together the articles as an additional fairness contribution degree (“+15” in FIG. 4). The display information generation unit 209 estimates the current overall fairness score using the fairness score transition indicated by the object 505 and the additional fairness contribution degree. Further, details of how the fairness has improved may be disclosed. The additional fairness contribution corresponds to the fairness score added by the accuracy determining section 205 and the fairness determining section 206. As already explained, the impartiality score added by the fact check is attributed to the expert, who is given the impartiality score.

Figure 11 shows the hardware configuration that implements this system. The CPU 413 starts the program of this system and executes the processing of the functional blocks shown in FIG. Data used for calculations is temporarily stored in memory 415. When high-speed calculation is required, data may be transferred to the memory of the GPU 414, and the GPU 414 may support the processing performed by each functional block. The display information generated by this system is displayed on the display device 411, and can be previewed by the system operator. A database such as article DB 3001 that stores news articles is preferably stored on the storage device 412. This system may be configured with a plurality of computers via the communication interface 417, and actual calculation processing may be executed on an instance on the cloud.

This server may be provided as an on-premises server or may be implemented on a public cloud. When implementing in a public cloud, a configuration may be used in which the CPU 413, GPU 414, and memory 415 are dynamically allocated and scaled according to the calculation scale. The system (fairness management unit) may be implemented using the same hardware as the system (back end and front end), or may be implemented using separate hardware. Here, we assume a split implementation, and indicate it by adding a -A sign at the end.

It is preferable that the fairness reward distribution section 210 realizes its functions by using the fairness score management section 211 and the display devices 411-A to 417-A. In particular, communication between the system (back end and front end) and the system (fairness management unit) is performed through communication I/Fs 417 and 417-A, which constitute a blockchain network N and exchange data blocks. The storage device 421-A of the system (fairness management department) maintains a fairness management ledger for each news article and a management ledger of fairness contribution for each user, and transactions via the blockchain It is updated by approving the transaction regarding the increase/decrease value of the fairness contribution, which is output to the blockchain by the fairness judgment unit in (and front end).

The communication I/F 417 is used to transfer the news viewing website GUI from the server to the user. The user terminal also has a similar hardware configuration. If the terminal is hardware with a web browsing function, such as a personal computer, smartphone, or game console, then a terminal with a similar hardware configuration may be used as a network participation terminal of this system.

With the above, it is possible to provide a mechanism that allows you to earn rewards in the form of tokens when you contribute to the fairness of news articles. When users post comments on SNS or news sites, this system collects the information and processes it to make it fair.

2. Second Embodiment In the first embodiment, the case where the system actively retrieves data has been described. In the second embodiment, a case will be described in which data is collected passively from the system's perspective, in which the user provides information or notifies the system of information regarding the user's reaction.

Specifically, in the first embodiment, as described with respect to the operation on the interface 508, the fairness information UI 208 allows information to be provided through the information providing interface 508. The processing in this case will be explained below.

For example, the accuracy determination unit 205 requires primary source information to ensure the accuracy of the information in the original news article of the buzz news, but as shown in FIG. 13, the accuracy determination unit 205 After receiving information from the user via the fairness information UI 208 (S1301), it is determined whether or not something useful has been obtained regarding the primary source (S1302).

The determination of whether or not the information is useful as a primary source may be performed automatically by the system or manually. For example, the accuracy determining unit 205 determines whether the primary source to be determined is a government-approved reference document, and if it determines that the primary source is a government-approved reference document, the accuracy determination unit 205 As in the case above, the primary source is analyzed (S1303).

As a result of analyzing the primary source, the accuracy determining unit 205 determines whether the content of the news article has been altered (S1304). If the accuracy determination unit 205 determines that the content of the news article has not been altered (S1304; No), it adds the fairness score of the news article (S1305). If the accuracy determination unit 205 determines that the content of the news article has not been altered (S1304; Yes), it reduces the fairness score of the news article (S1306). In response, the accuracy determination unit 205 recalculates and updates the reporter's fairness score (S1307, S1308), and also assigns a fairness score to the information provider (S1309).

The original data for generating opinion clusters, which is the original data for the fairness determination unit 206, may be input directly from the information providing interface 508. In this system, fairness is defined as presenting a variety of opinions and visualizing the position among them, so even if many extreme opinions are posted, there will be no influence.

In this case, if there is a change in the fairness score of the article as a result of this processing, the fairness reward distribution unit 210 assigns a degree of fairness contribution to the user who played a role in forming the opinion cluster. By creating a system that distributes rewards based on the degree of fairness contribution among clusters of the same opinion, the system can be made robust against opinion manipulation by some users who post a large number of posts.

Furthermore, we have explained how the system of the present invention selects experts for the fact check unit 207 from the expert database and the fairness contribution ledger for each user. A method of directly providing fact check information using the providing interface 508 is also conceivable. However, self-nominated fact checks will only be allowed if they have a proven track record in the relevant news field. A case with a proven track record is assumed to be a case where the paper has been certified by a third-party organization, such as a past peer-reviewed paper accepted by an authoritative academic society. As described above, it is a good idea to provide a mechanism to enhance the fairness of news by using information posted by users to augment information that the system could not collect. This system has been explained above using drawings, but as explained using FIG. A news distribution system that distributes news, which monitors news sites on the network and allows quotations or comments on news articles published on the news sites to be posted under predetermined conditions (for example, the number of citations or comments on a news article). a buzz news detection unit 201 that detects buzz news that satisfies the following criteria (Is the number of buzz news items greater than or equal to a predetermined threshold value?), and a comment on the buzz news that is written on the network in association with the buzz news and the comment. The sensitivity information collection unit 202 collects transaction data including the cumulative value of If a condition (for example, the number of clusters is greater than or equal to a predetermined threshold) is satisfied, the sensitivity analysis unit 203 determines that the news article with the comment is a highly fair news article, and the sensitivity analysis unit 203 determines that the news article with the comment is a highly fair news article A related article collection unit 204 that collects related articles including primary sources before media editing for the article, the news articles determined by the sensitivity analysis unit, and the related articles collected by the related article collection unit. an accuracy determination unit 205 that determines the accuracy of the news article based on the following; the fairness of the news article determined by the sensitivity analysis unit; and the accuracy of the news article determined by the accuracy determination unit. a fairness determination unit 206 that calculates a fairness score that is a numerical value; a fairness management ledger 801 that associates the fairness score with the news article; a fairness score management unit 211 that manages a fairness contribution management ledger 9101 that associates the fairness contribution degree, which is a fairness score given to the user when the user It includes a fairness reward distribution unit 210 that uses the management ledger and the fairness contribution degree management ledger to give a reward to a user who has taken an action to increase the fairness of the news article. This ensures the transparency of the position of reporters and other senders of news articles, making it possible to deliver accurate and highly fair news.

Further, as explained in FIG. 2, S301, etc., the buzz news detection unit detects the buzz news when the number of citations or the number of comments of the news article reaches a predetermined threshold as the predetermined condition. Therefore, fairness can be judged based on the number of citations and comments on a news article.

Further, as explained in FIG. 5, S506, etc., the accuracy determination unit compares the information of the primary source with the gist of the news article, and if there is no change in content, adds the fairness score. . This makes it possible to confirm that an accurate news article is being sent without compromising the gist of the original news article.

Further, as explained in FIG. 5, S502, etc., the accuracy determination unit determines that there is a description disclosing the primary source in the news article, and that the primary source can actually be accessed and viewed. If so, add the fairness score of the news article. Thereby, it is possible to ensure the objectivity that the news article is based on the primary source, and it is possible to calculate a fairness score that takes this objectivity into account.

In addition, as explained in FIG. 5, S509, etc., a predetermined expert is requested to fact check the above news article, and if the result of the fact check is that the content of the above news article is positive, the above It has a fact check unit 207 that adds the fairness score of the news article. This can ensure that the content of the news article is fair.

In addition, as explained using FIG. 7 etc., the fairness determination unit calculates a fairness score by adding a higher fairness score as the number of opinion clusters increases, so that more diverse opinions are taken into consideration. You can confirm that the news article is highly fair.

In addition, as explained using FIG. Since the system includes a display information generation unit 209 that displays a fairness score on the computer screen, a user of this system can easily check the current fairness score for a news article.

Further, as explained using FIG. 4 and the like, the display information generation unit displays a graph showing the temporal transition from the raw score of the news article to the overall fairness score on the screen. Since the information is displayed, the user of this system can easily check the change in the fairness score over time.

Further, as explained using FIG. 4 and the like, the display information generation unit reads the fairness contribution management ledger and displays the breakdown of users who contributed to the increase/decrease in the fairness contribution on the screen. Therefore, the user of this system can easily check the users who have contributed to the increase/decrease in the fairness contribution degree.

Further, as explained using FIG. 4 etc., the display information generation unit displays on the screen an interface that provides information that contributes to improving fairness from the user,
Furthermore, as explained using FIG. 4 and the like, the accuracy determination unit determines the accuracy of the news article using the primary source obtained through the interface, so the accuracy determination unit determines the accuracy of the news article using the primary source obtained through the interface. Accuracy can be determined by the primary source.

Further, as explained using FIG. 10 etc., the fairness score management unit performs processing related to increase/decrease in the fairness contribution (for example, provision of primary sources, donations, fact checks) via the network. is received from a user, the user is associated with the news article targeted for the fairness contribution, and the amount of increase/decrease, and managed as a token exchanged on the blockchain. This allows the fairness contribution degree to be issued using tokens that circulate on the blockchain.

Further, as explained using FIG. 10 etc., the token has a transaction ID determined for each process related to the increase/decrease, and the fairness score management unit manages the fairness score for each transaction ID. Since it manages increases and decreases in the fairness score, it is possible to manage increases and decreases in the fairness score each time a user performs an action such as providing a primary source, making a donation, or checking facts.

Further, as explained using FIG. 10 etc., the fairness reward distribution unit gives the reward based on the fairness score that increases or decreases for each token identified by the transaction ID, so that the user , it is possible to give a reward according to a fairness score that increases or decreases each time the above action is performed.

Further, as explained using FIG. 10 etc., the reward distribution unit may manage the token in a format that allows for the amount of increase/decrease in the fairness contribution and exchange with other currencies. . Thereby, the degree of fairness contribution can be reflected as the value of other currencies.

In this way, this system can provide an ecosystem for distributing fair news. Furthermore, since this system uses a blockchain framework, it is possible to track the ownership and rights of data used in news articles, ensuring reliability and transparency. In addition, the rights themselves can be bought and sold through a marketplace.

The present invention is not limited to the above-described embodiments as they are, and in the implementation stage, components may be modified and embodied without departing from the gist thereof, or multiple components disclosed in the above-described embodiments may be used. These can be implemented in appropriate combinations.

201 Buzz news detection section 202 Emotional information collection section 203 Emotional analysis section 204 Related article collection section 205 Accuracy judgment section 206 Fairness judgment section 207 Fact check section 208 Fairness information user interface (UI)
209 Display information generation unit 210 Fairness reward distribution unit 211 Fairness score management unit 301 Buzz news detection 302 Reaction collection 303 Sensitivity analysis processing 304 News data collection 305 Fact check request 306 Submission of corrected article 307 Fact check result 308 Article accuracy Fairness/fairness determination processing 309 Creation/update of reference information source 310 Provision of reference information source site 311 Donation to reference information source 312 Token distribution 313 User fairness contribution update processing 411 Display device 412 Storage device 413 CPU
414 GPUs
415 Memory 416 User interface 417 Communication interface 501 Verification information 502 Sensitivity analysis information 503 Fairness score raw score of original article 504 Fact check progress 505 Time-series changes in fairness score 506 User actions that affected fairness score History 507 Donation reception interface 508 Information reception interface 509 Backlog reference interface

Claims (15)

A news distribution system that distributes predetermined news by a computer executing a program,
a buzz news detection unit that monitors news sites on a network and detects buzz news whose quotes or comments on news articles published on the news sites meet predetermined conditions;
a sensitivity information collection unit that collects transaction data including comments on the buzz news written on the network in association with the buzz news and cumulative values of the comments;
If the number of clusters obtained by analyzing the comments included in the transaction data and clustering the opinions included in the comments satisfies a predetermined condition, the news article to which the comments are attached is considered to be unbiased. The sensitivity analysis department determines that it is a high-quality news article,
a related article collection unit that collects related articles from the network, including primary sources for the news article before being edited by the media;
an accuracy determination unit that determines the accuracy of the news article based on the news article determined by the sensitivity analysis unit and the related articles collected by the related article collection unit;
a fairness determination unit that calculates a fairness score that quantifies the fairness of the news article determined by the sensitivity analysis unit and the accuracy of the news article determined by the accuracy determination unit;
A fairness management ledger that associates the fairness score with the news article, and a fairness contribution degree that is a fairness score given to the user when the user performs an action that contributes to an increase or decrease in the fairness score. and a fairness score management ledger that associates the user with the fairness score management ledger;
a fairness reward distribution unit that uses the fairness management ledger and the fairness contribution management ledger to reward a user who has taken an action to increase fairness with respect to the news article;
A news distribution system characterized by having. The news distribution system according to claim 1,
The buzz news detection unit detects the buzz news when the number of citations or the number of comments of the news article reaches a predetermined threshold as the predetermined condition.
A news distribution system characterized by: The news distribution system according to claim 1,
The accuracy determination unit compares the information of the primary source and the gist of the news article, and adds the fairness score if the content has not been altered.
A news distribution system characterized by: The news distribution system according to claim 1,
The accuracy determining unit adds a fairness score to the news article if there is a description disclosing a primary source in the news article and the primary source can actually be accessed and viewed.
A news distribution system characterized by: The news distribution system according to claim 1,
a fact check unit that requests a predetermined expert to fact check the news article, and adds a fairness score of the news article if the result of the fact check is a positive evaluation of the content of the news article;
A news distribution system characterized by having. The news distribution system according to claim 1,
The fairness determination unit calculates a fairness score by adding a higher fairness score as the number of opinion clusters increases.
A news distribution system characterized by: The news distribution system according to claim 1,
A current overall fairness score is calculated based on the raw score of the news article and the increase/decrease in the fairness contribution, and the calculated overall fairness score is displayed on a computer screen. display information generation unit;
A news distribution system characterized by having. The news distribution system according to claim 7,
The display information generation unit displays, on the screen, a graph showing a temporal transition from the raw score of the news article to the overall fairness score.
A news distribution system characterized by: The news distribution system according to claim 7,
The display information generation unit reads the fairness contribution management ledger and displays on the screen a breakdown of users who have contributed to an increase or decrease in the fairness contribution.
A news distribution system characterized by: The news distribution system according to claim 1,
The display information generation unit displays on the screen an interface that provides information that contributes to improving fairness from a user,
The accuracy determination unit determines the accuracy of the news article using the primary source obtained via the interface.
A news distribution system characterized by: The news distribution system according to claim 1,
The fairness score management unit receives a process related to an increase or decrease in the fairness contribution degree from a user via the network, and includes the user, the news article targeted for the fairness contribution degree, and the amount of increase or decrease. and manage them as tokens exchanged on the blockchain.
A news distribution system characterized by: The news distribution system according to claim 11,
The token has a transaction ID determined for each process related to the increase/decrease, and the fairness score management unit manages the increase/decrease in the fairness score for each transaction ID.
A news distribution system characterized by: The news distribution system according to claim 11,
The fairness reward distribution unit gives the reward based on the fairness score that increases or decreases for each of the tokens identified by the transaction ID.
A news distribution system characterized by: The news distribution system according to claim 13,
The reward distribution unit manages the tokens according to the amount of increase/decrease in the fairness contribution and in a format that can be exchanged with other currencies.
A news distribution system characterized by: A news distribution method performed in a news distribution system that distributes predetermined news by a computer having a processor and a memory, the method comprising:
Monitor news sites on the network, detect buzz news where quotations or comments on news articles published on the news site meet predetermined conditions,
Collecting transaction data including comments on the buzz news written on the network in association with the buzz news and cumulative values of the comments;
If the number of clusters obtained by analyzing the comments included in the transaction data and clustering the opinions included in the comments satisfies a predetermined condition, the news article to which the comments are attached is considered to be unbiased. It is determined that it is a high-quality news article,
Collecting related articles from the network, including primary sources of the news article before it is edited by the media;
determining the accuracy of the news article based on the determined news article and the collected related articles;
Calculating a fairness score that quantifies the determined fairness of the news article and the determined accuracy of the news article,
A fairness management ledger that associates the fairness score with the news article, and a fairness contribution degree that is a fairness score given to the user when the user performs an action that contributes to an increase or decrease in the fairness score. and a fairness contribution management ledger that associates the user with the user;
using the fairness management ledger and the fairness contribution management ledger to reward a user who has taken an action to increase the fairness of the news article;
A news distribution method characterized by having the following.

Images