KR102242317B1 - Qualitative system for determining fake news, qualitative method for determining fake news, and computer-readable medium having a program recorded therein for executing the same - Google Patents

Abstract

The qualitative fake news determination system of the present invention includes a user terminal and a server capable of communicating with the user terminal, and the server uses a data scraping technology to provide real news, fake news, and sources of each news. An information collection unit that collects news information including information; An estimating unit for grouping the news based on the similarity between the news and estimating a sharing pattern by analyzing the source information for each group; A probability value calculating unit that predicts a proportion of the fake news for each group based on the sharing pattern and the source information; A data set generator configured to generate a training data set by classifying each news into one or more classes based on the ratio; A database storing the training data set; An artificial intelligence processing device for generating an artificial intelligence model by performing machine learning based on the training data set; And a request input unit that receives a request for determining fake news from the user terminal, wherein the artificial intelligence model is used to calculate a probability that the news displayed on the user terminal is fake, according to which, In the process of creating, sharing, and disseminating news, you can understand the impact of fake news on the Internet ecosystem, while providing insights on how to block or reduce fake news, and judge fake news.

Description

Qualitative system for determining fake news, qualitative method for determining fake news, and computer-readable medium having a program recorded therein for executing the same}

The present invention relates to a qualitative fake news determination system, a determination method, and a computer-readable recording medium recording a program for executing the same, and more particularly, to qualitatively detect fake news based on the facts, values, and context of the news. It relates to a system, a judgment method, and a recording medium using a natural language processing algorithm among artificial intelligence element technologies for discrimination.

The police define fake news as'information disseminated in a way that improves credibility by processing it to look like actual media coverage'. It is defined as'information'.

Anyone can make fake news, and internet sites that help make it are overflowing, and recently, single-person broadcasts and YouTubers function like broadcasts, and the frequency of fake news is increasing. DailyPadak provides a service that makes any content look like an article if you put a title and reporter's name like an article and attach a photo. It is not illegal to help make fake news or make fake news at the level of pranks, the Election Commission explained. The Election Commission approached fake news in terms of freedom of expression, and as such, the concept, definition, and scope of fake news are quite confused.

According to the Korea Press Foundation's Fake News Status and Problems (2017) report, 76.3% of respondents received fake news through Internet services such as portals, Facebook, and KakaoTalk. It seems that the vast majority of the distribution channels of fake news are Internet services. Only a few were exposed to media such as newspapers and TV at 9.1%, and private gatherings such as friends or seniors and juniors at 7.7%.

In particular, among them, 39.7% of users who received fake news through mobile messengers such as KakaoTalk and Line were the highest. What stands out is that the proportion was particularly high at 45.6% in their 50s. As the election season approaches, fake news about specific candidates is distributed through Kakao Talk to the elderly, who are the information-vulnerable class. The'fare discount' Chirashitek level is at the level of'Aegyo', but the story gets serious when it goes to spreading false facts for the purpose of slander.

Social platforms such as Facebook and Twitter also recorded a not small percentage at 27.7%. Internet cafes/communities and blogs were also high at 24.3%. YouTube and Africa TV followed 4.6%, followed by fake news sites 3.7%.

The media foundation said, “Online users exchange texts and photos about specific or false facts in single chat rooms, and this quickly spreads out and becomes fake news.” “The distribution of fake news is through mobile messengers and social platforms. It is being mediated and spreading through content distribution platforms such as,” he analyzed.

The powerful destructive power of fake news lies in this distribution structure. Chirashi is mainly distributed through group katokbangs, while fake news is publicly distributed and spread. Fake news is mainly flooded with Internet communities such as Daily Best Store (Ilbe), Today's Humor (Oyu), and Pompu. If the site is well received, it will be spread to SNS (social network services/sites) such as Facebook, KakaoTalk, Twitter, and Naver Band.

The problem is that even if such fake news spreads, there is no way to stop it.

In the case of acquaintance-based services such as KakaoTalk and Facebook, even if they receive fake news, the reliability of the information tends to be higher than that of the general case, as the'acquaintance' delivered it. In addition, due to the nature of the Internet service, which spreads quickly, even if the victim reports to the police and investigates the spreader, the fact of the damage is already snowballed in most cases.

Accordingly, there is a need for a tool that can provide insight into the reliability of online content by identifying reliable news sources.

As a prior art, there is a domestic registered patent No. 10-1864439 (Patent Document 1). Patent Document 1 provides a fake news detection system with a graphic user interface screen window capable of identifying fake news.

According to this, by separating the comments of True and Fake, the environment is created so that they can argue logically with each other, rather than an emotional fight, and the collective self-defense of netizens by writing links or postings on articles or rumors in question. Can be used to compare-judge.

However, this is ultimately discriminated based on the opinions of the majority of netizens, which is difficult to see as a system capable of clearly discriminating fake news, and the lower the participation, the lower the discrimination power. On the other hand, there is an expert-based system for reporters, media experts, and journalists to judge whether actual news is true or fake. However, at the present time when fake news is expanding quantitatively, it is impossible to distinguish between true and fake by hand by limited experts.

Therefore, at present, regardless of the opinions of netizens, a fake news discrimination system is needed to determine whether the value, content, and context of the news are true or false with an algorithm based on natural language processing learned from real news with artificial intelligence technology. .

KR 10-1864439 B1

An object of the present invention for solving the conventional problems as described above is to provide a system and method capable of blocking or reducing fake news by qualitatively discriminating fake news by grasping the motives of producing and sharing fake news.

The qualitative fake news determination system of the present invention for achieving the above object includes a user terminal and a server capable of communicating with the user terminal, and the server uses a data scrapping technology to provide real news and fake news. , And an information collection unit for collecting news information including source information of each news; An estimating unit for grouping the news based on the similarity between the news and estimating a sharing pattern by analyzing the source information for each group; A probability value calculating unit that predicts a proportion of the fake news for each group based on the sharing pattern and the source information; A data set generator configured to generate a training data set by classifying each news into one or more classes based on the ratio; A database storing the training data set; An artificial intelligence processing device for generating an artificial intelligence model by performing machine learning based on the training data set; And a request input unit receiving a request for determining fake news from the user terminal, wherein the artificial intelligence model is used to calculate a probability that the news displayed on the user terminal is fake.

In addition, the similarity is preferably calculated by comparing at least one of the title, main sentence, field, and production time of each of the news.

In addition, the sharing pattern is preferably estimated by using at least one of the number of news included in the group, the type of news media, and the density of production time.

In addition, the sharing pattern is preferably estimated by determining the order of the news in chronological order based on the production timing of the news included in the group, and calculating the spreading path and speed of the news.

In addition, the main sentence is at least one of a morpheme analysis step, an entity name recognition step, a syntax analysis step, a proxy processing step, a vocabulary semantic analysis step, a meaning inverse recognition step, and a cross-reference step for the news information by a processor. It is preferably extracted by performing one step.

In addition, when the probability value calculation unit predicts the ratio, it is preferable that the ratio is changed according to whether the title of each news and the content of the main sentence are inconsistent.

In addition, when predicting the ratio in the probability value calculation unit, it is preferable that the ratio is changed according to whether the context of each news is inconsistent.

In addition, the artificial intelligence processing apparatus stores an algorithm used for artificial intelligence, and the algorithm is preferably an artificial neural network or a fuzzy neural network.

The qualitative fake news determination method of the present invention for achieving the above object is, by the information collection unit included in the server, data scrapping technology is used to provide real news, fake news, and source information of each news. Collecting news information including; Estimating a sharing pattern by grouping the news by an estimation unit included in the server based on the similarity between the news and analyzing the source information for each group; Predicting a proportion of the fake news for each group based on the sharing pattern and the source information by a probability value calculator included in the server; Generating a training data set by classifying each news into one or more classes based on the ratio by a data set generator included in the server; Generating an artificial intelligence model by performing machine learning based on the training data set by an artificial intelligence processing device included in the server; And calculating a probability that the news displayed on the user terminal is fake by using the artificial intelligence model.

In addition, the calculating the probability that the news is fake may include inputting the news displayed on the user terminal to the server; Collecting, by the information collecting unit, news information similar to the news by using a data scrapping technique; Estimating a sharing pattern based on the collected similar news information; Determining whether the news is fake by using the artificial intelligence model and calculating the probability based on the sharing pattern; And outputting one of the probability of the news and whether the news is fake as a result.

Furthermore, if the probability that the other news is fake among other news similar to the news is more than a preset probability, it may be listed and output as a result.

In order to achieve the above object, the present invention further provides a computer-readable recording medium in which a program for executing a qualitative fake news determination method is recorded.

Details of other embodiments are included in the detailed description and the accompanying drawings.

Advantages and/or features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in a variety of different forms, only this embodiment is intended to complete the disclosure of the present invention, and the general knowledge in the technical field to which the present invention belongs. It is provided to fully inform the person of the scope of the present invention.

As described above, according to the present invention, in the process of creating, sharing, and distributing fake news, it is possible to determine whether or not it is fake news, thereby providing a method of blocking or reducing fake news.

1 is a conceptual diagram for explaining the configuration of a qualitative fake news determination system of the present invention.
2 is a conceptual diagram illustrating operations in a server of the qualitative fake news determination system of the present invention.
3 is a diagram illustrating an example of outputting a result on a user terminal of the qualitative fake news determination system of the present invention.
4 is a flowchart illustrating a method of determining qualitative fake news according to the present invention.
5 is a flowchart for further explaining the method of determining qualitative fake news according to the present invention.

Before describing the present invention in detail, terms or words used in the present specification should not be interpreted as being unconditionally limited to their usual or dictionary meanings, and in order for the inventors of the present invention to describe their invention in the best way It should be understood that the concepts of various terms can be appropriately defined and used, and furthermore, these terms or words should be interpreted as meanings and concepts consistent with the technical idea of the present invention.

That is, the terms used in the present specification are only used to describe the preferred embodiments of the present invention, and are not intended to specifically limit the content of the present invention, and these terms are used to describe various possibilities of the present invention. It should be noted that this is a term defined in consideration.

In addition, in this specification, it should be understood that the singular expression may include a plural expression unless clearly indicated in a different meaning in the context, and even if similarly expressed in the plural, the singular expression may include the meaning of the singular number. do.

Throughout the present specification, when a component is described as "including" another component, it does not exclude any other component, but further includes any other component unless otherwise indicated. It can mean that you can do it.

Furthermore, in the case where a component is described as "existing inside or connected and installed" of another component, the component may be directly connected to or installed in contact with the other component, and It may be installed spaced apart by a distance, and in the case of installation spaced apart by a certain distance, a third component or means may exist for fixing or connecting the component to other components. It should be noted that a description of the elements or means of 3 may be omitted.

On the other hand, when a component is described as being "directly connected" or "directly connected" to another component, it should be understood that there is no third component or means.

Likewise, other expressions describing the relationship between each component, such as "between" and "directly between", or "neighbor to" and "directly neighbor to" have the same effect. Should be interpreted as.

In addition, in the present specification, terms such as “one side”, “the other side”, “one side”, “the other side”, “first”, and “second”, if used, refer to one constituent element for one constituent element. Is used to be clearly distinguishable from other constituent elements, and it should be noted that the meaning of the constituent element is not limitedly used by such terms.

In addition, terms related to positions such as "upper", "lower", "left", and "right" in the present specification, if used, should be understood as indicating a relative position in the drawing with respect to the corresponding component, These position-related terms should not be understood as referring to absolute positions unless absolute positions are specified for their positions.

Furthermore, in the specification of the present invention, terms such as "... unit", "... group", "module", "device", if used, mean a unit capable of processing one or more functions or operations, which is a hardware Or it should be noted that it can be implemented in software, or a combination of hardware and software.

In addition, in the present specification, in specifying the reference numerals for each component of each drawing, the same reference numerals for the same components even if the components are indicated in different drawings, that is, the same reference throughout the specification. The symbols indicate the same components.

In the drawings attached to the present specification, the size, position, coupling relationship, etc. of each component constituting the present invention are partially exaggerated, reduced, or omitted in order to sufficiently clearly convey the spirit of the present invention or for convenience of description. It may have been described, and therefore its proportion or scale may not be exact.

In addition, in the following description of the present invention, a detailed description of a configuration determined to unnecessarily obscure the subject matter of the present invention, for example, a known technology including the prior art may be omitted.

Hereinafter, a qualitative fake news determination system, a determination method, and a computer-readable recording medium in which a program for executing the same according to an embodiment of the present invention is recorded will be described.

First, a qualitative fake news determination system according to an embodiment of the present invention will be described with reference to FIGS. 1 to 3.

1 is a conceptual diagram for explaining the configuration of a qualitative fake news determination system of the present invention.

As shown in FIG. 1, the qualitative fake news determination system of the present invention includes a server 100, a communication network 200, and a user terminal 300.

The server 100 includes an information collection unit 110, an estimation unit 120, a probability value calculation unit 130, a data set generation unit 140, a database 150, an artificial intelligence processing unit 160, and a request input unit 170. ) And an output unit 180 is preferably included. Each unit of the server 100 is preferably provided with one or more processors.

The information collection unit 110 is preferably a unit that collects news information including real news, fake news, and source information of each news by using a data scraping technology in the server 100.

Using data scraping technology, various structured, unstructured, and semi-structured data exposed to web browsers or SNS can be analyzed in real time. Data scraping technology is a technology that automatically connects to a system, displays data on the screen, and extracts and retrieves necessary data. Web scraping is a technology that extracts information from a specific part of a website. Through this, information of a website or program can be retrieved and stored in another program or DB so that it can be viewed and used whenever necessary, and the stored data can be used as comparative analysis data.

Here, the source information may include a website on which the corresponding news was first produced, a creator, a production date and time, a type of news media, a distribution channel, a consumer, and the like.

The estimating unit 120 is preferably a unit connected to the information collecting unit 110 to estimate a sharing pattern by grouping the news based on the similarity between the news and analyzing source information for each group.

An example of calculating grouping, sharing patterns, and probability values will be described with further reference to FIG. 2.

2 is a conceptual diagram illustrating operations in a server of the qualitative fake news determination system of the present invention.

As shown in FIG. 2, similar news 1a, news 1b, news 1c, and the like may be grouped into group 1. As shown in FIG.

Here, the similarity is preferably calculated by comparing at least one of the title, main sentence, field, and production time of each news. For example, among economic news, news of similar topics created at a similar time may be grouped into a group. Here, the title, main sentence, and field of the news may be analyzed to determine whether it is a similar subject.

Here, the main sentence is at least one of a morpheme analysis step, an entity name recognition step, a syntax analysis step, a proxy processing step, a vocabulary semantic analysis step, a meaning inverse recognition step, and a cross-reference step for news information by the processor. It is preferably extracted by carrying out.

The sharing pattern is preferably estimated by using at least one of the number of news included in the group, the type of news media, and the density of production time.

For example, there are 25 news stories with high similarity in one group, and among them, Facebook, foreign media, blogs, and KakaoTalk, all of which were produced and distributed within two days can be analyzed. Based on this, the estimating unit can calculate the path and speed of spreading the news. As described above, the sharing pattern is preferably estimated by determining the order of the news in chronological order based on the production timing of the news included in the group, and calculating the spreading path and speed of the news.

As shown in FIG. 2, sources of news 1a grouped in group 1 may be identified as source 1, source 2, and the like. At this time, the source can be identified as a medium such as a blog type or a press company type.

As shown in FIG. 2, a probability value may be calculated according to the number of appearances of each medium and the production date and time of each news.

As shown in FIG. 1, it is preferable that the probability value calculation unit 130 is connected to the estimation unit 120 and predicts a ratio, that is, a probability value, of fake news for each group based on the sharing pattern and source information. For such a probability value, the accuracy may gradually increase by repetitive learning and correction through artificial intelligence.

When the probability value calculation unit 130 predicts the proportion of fake news, when only the sharing pattern and source information is analyzed, the news has spread mainly to report media that are less reliable, or additionally through SNS, the spread rate (normal activity If a blogger who didn't do it suddenly continuously shares related news, or signs that a large number of people are spreading certain news systematically) exceeds a preset level, the probability of being faked increases.

In this case, the title and content of the main sentence of the news are additionally analyzed, and if the title and the content are inconsistent, the probability of being fake may be higher. In addition, in another embodiment, when the probability value calculator 130 predicts the proportion of fake news, the ratio may be determined according to whether the context of each news is inconsistent. For example, if there is a sentence that is not related to the context in the contents of the news (the news is made abstract among several methods, and the distant sentence is selected as a sentence that is not related to the context while comparing the summary with the text), the news is It can be more likely to be fake.

Or, it is possible to derive more sophisticated values by processing inconsistencies between the title and contents of the news, and inconsistencies in specific sentences or paragraphs in the news contents, respectively, or by changing the order of sequential processing.

The dataset generation unit 140 is preferably a unit that is connected to the probability value calculation unit 130 and generates a training dataset by classifying each news into one or more classes based on the ratio.

The database 150 stores the training data set generated by the data set generation unit 140. The database 150 may be connected to not only the training dataset, but also the estimating unit 120, the probability value calculating unit 130, and the artificial intelligence processing apparatus 160, and may store data from each of them.

That is, the database 150 may store a shared pattern in the estimation unit 120, a ratio in the probability value calculation unit 130, and the artificial intelligence model 161 generated by the artificial intelligence processing apparatus 160.

The artificial intelligence processing apparatus 160 is preferably a unit that generates an artificial intelligence model 161 by performing machine learning based on a training data set.

Algorithms used for artificial intelligence are stored in the artificial intelligence processing device 160, and it is preferable to use artificial neural networks or fuzzy neural networks as the algorithm. In particular, it is preferable that the artificial intelligence processing apparatus 160 stores an algorithm for processing natural language.

Artificial neural networks are classified into several types according to their structure and function, and the most common artificial neural network is a multilayer perceptron with multiple hidden layers between one input layer and an output layer. The artificial neural network can be implemented through the artificial intelligence processing device 160, and is a form in which several neurons, which are basic computing units, are connected by a weighted link, so that the weighted link can adapt to a given environment. You can adjust the weight.

The artificial neural network includes various models such as a self-organizing map (SOM), a recurrent neural network (RNN), and a convolutional neural network (CNN), but is not limited thereto.

Fuzzy neural networks are systems that use the learning capabilities of neural networks to express rules verbally or update new rules in a knowledge base. In the case of using a fuzzy neural network in the present invention, rules may be defined and used to calculate a shared pattern, similarity, and/or a false probability value.

The request input unit 170 is preferably a unit that receives a request for determining fake news from a user terminal. Here, the fake news determination request is a request received from the user terminal 300, and is preferably information including the title, body, source information, reporter information, sharing frequency and spread information per minute of the news to be determined.

The user terminal 300 is preferably a user terminal such as a computer equipped with a smart phone, a tablet, a touch screen, and a smart watch, which is a kind of wearable.

In this way, when a request is input to the request input unit 170, the artificial intelligence model 161 generated by the artificial intelligence processing device 160 and stored in the database 150 is used, and the news displayed on the user terminal 300 is fake. Probability can be calculated.

The calculated probability value may be output through the output unit 180 and transmitted from the server 100 to the user terminal 300 through the communication network 200.

As shown in FIG. 3, a probability value may be displayed on the terminal 300. 3 is a diagram illustrating an example of outputting a result on a user terminal of the qualitative fake news determination system of the present invention.

As shown in FIG. 3, the process until the probability value is displayed on the user terminal 300 will be described in more detail starting from the operation in the user terminal 300.

First, news A (or news information including source information) through the user terminal 300 is received by the server 100 through the communication network 200 and input to the request input unit 170.

The information collection unit 110 collects information similar to news A through data scraping, for example, news A1, news A2, news A3, etc., and sharing pattern A in the estimation unit 120 based on the collected information. Estimate

When the sharing pattern for news is estimated through the estimating unit 120, the artificial intelligence model 161 stored in the database 150 is used, so that a probability value can be immediately calculated using only the sharing pattern. That is, a probability value for another sharing pattern similar to the sharing pattern A is applied, the probability value is output through the output unit 180 and displayed on the user terminal 300.

In another embodiment, not only the sharing pattern, but also whether the content of the title and main sentence of News A is inconsistent, or whether there is a context inconsistency in the body of News A may be analyzed and determined by the processor. Accordingly, the probability value calculation unit 130 may finally calculate a probability value from the artificial intelligence model 161 and a probability value from another processor through a predetermined equation.

The probability value finally calculated as described above may be transmitted to the user terminal 300 through the output unit 180 and displayed on the display of the user terminal 300.

Next, a method of determining qualitative fake news according to an exemplary embodiment of the present invention will be described with reference to FIG. 4 further.

4 is a flowchart illustrating a method of determining qualitative fake news according to the present invention.

As shown in FIG. 4, the method for determining qualitative fake news according to a preferred embodiment of the present invention includes an information collection step (S100), a news grouping step (S200), a sharing pattern estimation step (S300), and a probability value calculation step. (S400), a data set generation step (S500), and an artificial intelligence model generation step (S600).

In the information collection step (S100), a data scraping technology is used by the information collection unit 110 included in the server 100 to provide real news, fake news, and source information of each news, news production company. News information including information, news author (reporter) information, news production time information, and real-time issue information of the news is collected.

In the news grouping step (S200), news is grouped based on the similarity between each news by the estimation unit 120 included in the server 100.

In the sharing pattern estimation step (S300), it is preferable that the source information, reporter information, media company information, and news publishing time information are analyzed for each group by the estimation unit 120 to estimate the sharing pattern.

In the probability value calculation step S400, the percentage of fake news for each group is predicted based on the sharing pattern and source information by the probability value calculator 130 included in the server 100.

In the data set generation step (S500), each news is classified into one or more classes based on the ratio by the data set generation unit 140 included in the server 100 to generate a training data set. Data sets are trained by classifying them into validation, test, and learning purposes.

In the artificial intelligence model generation step (S600), the artificial intelligence model 161 is generated by performing machine learning based on the training dataset by the artificial intelligence processing device 160 included in the server 100.

With further reference to FIG. 5, until the artificial intelligence model 161 is used in the qualitative fake news determination method according to an exemplary embodiment of the present invention to calculate the probability that the news displayed on the user terminal 300 is fake. Explain the detailed steps.

5 is a flowchart for further explaining the method of determining qualitative fake news according to the present invention.

The step of calculating the probability that the news is fake includes the news input step (S710), the similar information collection step (S720), the sharing pattern estimation step (S730), the step of determining using an artificial intelligence model (S740), and the result output. It includes step S750.

In the news input step S710, news displayed on the user terminal 300 may be input to the server 100.

In the similar information collection step S720, the information collection unit 110 uses a data scraping technology to collect news information similar to the news.

In the sharing pattern estimation step S730, the sharing pattern may be estimated based on the collected similar news information by the estimating unit 120.

In the step of determining using the artificial intelligence model 161 (S740), the artificial intelligence model 161 is used and the probability is calculated based on the sharing pattern to determine whether the news is fake.

In the result output step S750, one of the probability of the news and whether the news is fake may be output as a result.

Further, in the result output step (S750), if the probability that the other news is fake among other news similar to the news is more than a preset probability, it may be listed and output as a result.

A computer-readable recording medium according to an embodiment of the present invention is characterized in that it is a computer-readable recording medium in which a program for executing a qualitative fake news determination method is recorded.

Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magnetic-optical media such as floppy disks, and ROM, RAM, flash memory, and the like. Hardware devices specially configured to store and execute the same program instructions are included. Examples of program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

As described above, according to the present invention, in the process of creating, sharing, and distributing fake news, it is possible to determine whether or not it is fake news, thereby providing a method of blocking or reducing fake news.

100: server
110: information collection unit
120: request input unit
130: output
140: data set generation unit
150: database
160: artificial intelligence processing device
161: artificial intelligence model
200: communication network
300: user terminal

Claims (17)

It includes a user terminal and a server capable of communicating with the user terminal,
The server,
An information collection unit that collects news information including real news, fake news, and source information of each news by using a data scraping technology;
An estimating unit for grouping the news based on the similarity between the news and estimating a sharing pattern by analyzing the source information for each group;
A probability value calculating unit that predicts a proportion of the fake news for each group based on the sharing pattern and the source information;
A data set generator configured to generate a training data set by classifying each news into one or more classes based on the ratio;
A database storing the training data set;
An artificial intelligence processing device for generating an artificial intelligence model by performing machine learning based on the training data set; And
Including; a request input unit for receiving a request to determine the fake news from the user terminal,
The artificial intelligence model is used to calculate the probability that the news displayed on the user terminal is fake,
The similarity is calculated by comparing at least one of the title, main sentence, field, and production time of each of the news,
The sharing pattern is estimated by using at least one of the number of news included in the group, the type of news media, and the density of production time,
The sharing pattern is estimated by determining the order of the news in chronological order based on the production time of the news included in the group, and calculating the spreading path and speed of the news,
When the news information is received to the server through the user terminal and input to the request input unit, a probability value for another sharing pattern similar to the sharing pattern is applied, and a probability value, which is the proportion of fake news, is output through the output unit. Characterized in that displayed on the,
Qualitative fake news judgment system.
delete delete delete The method of claim 1,
The main sentence is at least one of a morpheme analysis step, an entity name recognition step, a syntax analysis step, a proxy processing step, a vocabulary semantic analysis step, a meaning inverse recognition step, and a cross-reference step for the news information by a processor. Characterized in that it is extracted by performing,
Qualitative fake news judgment system.
The method of claim 5,
When predicting the ratio in the probability value calculation unit,
Characterized in that the ratio is changed according to whether the content of the title and the main sentence of each of the news are inconsistent,
Qualitative fake news judgment system.
The method of claim 5,
When predicting the ratio in the probability value calculation unit,
It characterized in that the ratio varies depending on whether the context of each news is inconsistent,
Qualitative fake news judgment system.
The method of claim 1,
Algorithms used for artificial intelligence are stored in the artificial intelligence processing device,
The algorithm is characterized in that it uses artificial neural networks or fuzzy neural networks,
Qualitative fake news judgment system.
Collecting news information including real news, fake news, and source information of each news by using a data scraping technology by an information collecting unit included in the server;
Estimating a sharing pattern by grouping the news by an estimation unit included in the server based on the similarity between the news and analyzing the source information for each group;
Predicting a proportion of the fake news for each group based on the sharing pattern and the source information by a probability value calculator included in the server;
Generating a training data set by classifying each news into one or more classes based on the ratio by a data set generator included in the server;
Generating an artificial intelligence model by performing machine learning based on the training data set by an artificial intelligence processing device included in the server; And
And calculating a probability that the news displayed on the user terminal is fake by using the artificial intelligence model,
The similarity is calculated by comparing at least one of the title, main sentence, field, and production time of each of the news,
The sharing pattern is estimated by using at least one of the number of news included in the group, the type of news media, and the density of production time,
The sharing pattern is estimated by determining the order of the news in chronological order based on the production time of the news included in the group, and calculating the spreading path and speed of the news,
When the news information is received to the server through the user terminal and input to the request input unit, a probability value for another sharing pattern similar to the sharing pattern is applied, and a probability value, which is the proportion of fake news, is output through the output unit. Characterized in that displayed on the,
How to judge qualitative fake news.
The method of claim 9,
The step of calculating the probability that the news is fake,
Inputting news displayed on the user terminal to the server;
Collecting, by the information collecting unit, news information similar to the news by using a data scraping technique;
Estimating a sharing pattern based on the collected similar news information;
Determining whether the news is fake by using the artificial intelligence model and calculating the probability based on the sharing pattern; And
And outputting one of the probability of the news and whether the news is fake as a result;
How to judge qualitative fake news.
delete delete delete The method of claim 9,
The main sentence is at least one of a morpheme analysis step, an entity name recognition step, a syntax analysis step, a proxy processing step, a vocabulary semantic analysis step, a meaning inverse recognition step, and a cross-reference step for the news information by a processor. Characterized in that it is extracted by performing,
How to judge qualitative fake news.
The method of claim 14,
When predicting the ratio in the probability value calculation unit,
Characterized in that the ratio is changed according to whether the content of the title and the main sentence of each news is inconsistent, or whether the context of each news is inconsistent,
How to judge qualitative fake news.
The method of claim 9,
Algorithms used for artificial intelligence are stored in the artificial intelligence processing device,
The algorithm is characterized in that it uses artificial neural networks or fuzzy neural networks,
How to judge qualitative fake news.
A computer-readable recording medium recording a program for executing the qualitative fake news determination method according to any one of claims 9, 10, 14 to 16.

Images