KR20190107830A - Spam-tag based blog spam detection apparatus and method, storage media storing the same - Google Patents

Abstract

The present invention relates to an apparatus for detecting blog spam based on spam tag, and to a method thereof. The apparatus for detecting blog spam based on spam tag comprises: a spam tag identification unit identifying a spam tag among a plurality of tags included in at least one post based on a body of the at least one post posted on a blog; a blog reliability calculation unit calculating the reliability of the blog based on the entire posts posted on the blog and a spam post including the spam tag; and a blog spam detection unit detecting the spam of the blog based on a tag spam degree corresponding to the reliability and the spam tag ratio of the spam post. Therefore, the apparatus can detect the blog spam by identifying spam tags included in the posts.

Description

Spam tag-based blog spam detection device and method, and recording media {SPAM-TAG BASED BLOG SPAM DETECTION APPARATUS AND METHOD, STORAGE MEDIA STORING THE SAME}

The present invention relates to a spam tag-based blog spam detection technology, and more particularly, to a spam tag-based blog spam detection device and method that can detect blog spam by including a spam tag included in a post.

It can be defined as representing a tag that does not match, that is, a tag that is not directly related to the subject of the article.

Korean Patent Registration No. 10-0902475 (2009.06.04) relates to a spam document determination system and method. The spam index of each spam detection technique is determined according to the accuracy rate of a plurality of spam detection techniques by continuously monitoring the spam document determination result. It can be assigned differently, and the accuracy rate can be increased along with the reproducibility of the spam document determination system.As a spam index is provided for each spam detection technique, even if a new spam detection technique emerges, only the spam index is assigned to the spam detection technique. It can also be applied immediately, increasing the scalability of the system.

Korea Patent Registration No. 10-0902475 (2009.06.04)

An embodiment of the present invention is to provide a spam tag-based blog spam detection device and method that can detect blog spam by identifying the spam tag included in the post.

An embodiment of the present invention is to provide a spam tag-based blog spam detection device and method that can calculate the reliability of the blog based on the spam post including the entire post and the spam tag posted on the blog.

An embodiment of the present invention is to provide a spam tag-based blog spam detection device and method that can detect whether a blog spam based on a weighted sum between blog reliability and tag spam.

Among the embodiments, the spam tag-based blog spam detection device identifies a spam tag among a plurality of tags included in the at least one post based on a body of at least one post posted to a blog. A spam tag identification unit, a blog reliability calculator that calculates the reliability of the blog based on the spam post including all the posts posted on the blog and the spam tag, and the reliability and the spam tag ratio of the spam posts It includes a blog spam detection for detecting whether the blog spam based on the tag spam degree.

The spam tag identification unit extracts a post key term through text parsing of the at least one post, and generates a post group by performing clustering based on the post key term. A tag analysis module for extracting a group key term for the post group and performing a comparative analysis between the group key term and a group tag included in the post group, and a spam tag identification module for identifying a spam tag using the result of the comparison analysis It may include.

The tag analysis module may derive the number of appearance frequencies, the number of appearance posts, and the appearance rate in the post group for each group tag through the comparative analysis.

The spam tag identification module may identify the group tag having the appearance ratio below a specific threshold value as a spam tag.

The blog reliability calculator may calculate a value obtained by dividing the number of spam posts by the total number of posts as the reliability of the blog.

The blog spam detector may determine the blog as spam when the weighted sum between the reliability and the tag spam degree exceeds a specific threshold.

Among the embodiments, the spam tag-based blog spam detection method may include (a) spam among a plurality of tags included in the at least one post based on a body of at least one post posted to a blog. Identifying a tag, (b) calculating the reliability of the blog based on a total post posted to the blog and a spam post including the spam tag, and (c) the reliability and spam tag rate of the spam post Detecting whether the blog is spam based on the degree of tag spam corresponding to the;

Step (a) is a step of generating a post group by extracting a post key term through text parsing of the at least one post and clustering based on the post key term. (a2) extracting a group key term for the post group, performing a comparative analysis between the group key term and a group tag included in the post group, and (a3) identifying a spam tag using the result of the comparative analysis It may include the step.

The step (a2) may be a step of deriving the number of appearances, the number of appearance posts, and the appearance rate in the post group for each group tag through the comparative analysis.

The step (a3) may be a step of identifying the group tag whose spam rate falls below a specific threshold value as a spam tag.

The step (b) may be a step of calculating a value obtained by dividing the number of spam posts by the total number of posts as the reliability of the blog.

Step (c) may be a step of determining the blog as spam when the weighted sum between the reliability and the tag spam degree exceeds a specific threshold.

In one or more embodiments, a computer-executable recording medium may include identifying a spam tag among a plurality of tags included in the at least one post based on a body of at least one post posted to a blog. Calculating the reliability of the blog based on the total posts posted on the blog and the spam post including the spam tag, and the tag spam level corresponding to the reliability and the spam tag ratio of the spam post. Detecting whether or not spam.

The disclosed technique can have the following effects. However, since a specific embodiment does not mean to include all of the following effects or only the following effects, it should not be understood that the scope of the disclosed technology is limited by this.

The spam tag-based blog spam detection apparatus and method according to an embodiment of the present invention may calculate the reliability of the blog based on the spam post including the entire post and the spam tag posted on the blog.

The spam tag-based blog spam detection apparatus and method according to an embodiment of the present invention can detect whether a blog is spam based on the weighted sum of blog reliability and tag spam.

1 is a diagram illustrating a spam tag-based blog spam detection system according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a blog spam detection device of FIG. 1.
3 is a block diagram illustrating a spam tag identification unit of FIG. 2.
4 is a flowchart illustrating a blog spam detection process performed by the blog spam detection device of FIG. 1.
FIG. 5 is a flowchart illustrating a spam tag identification process performed by the spam tag identification unit of FIG. 2.
6 is an exemplary view illustrating a process of identifying a spam tag performed by the spam tag identification unit of FIG. 2.

Description of the present invention is only an embodiment for structural or functional description, the scope of the present invention should not be construed as limited by the embodiments described in the text. That is, since the embodiments may be variously modified and may have various forms, the scope of the present invention should be understood to include equivalents capable of realizing the technical idea. In addition, the objects or effects presented in the present invention does not mean that a specific embodiment should include all or only such effects, the scope of the present invention should not be understood as being limited thereby.

On the other hand, the meaning of the terms described in the present application should be understood as follows.

Terms such as "first" and "second" are intended to distinguish one component from another component, and the scope of rights should not be limited by these terms. For example, the first component may be named a second component, and similarly, the second component may also be named a first component.

When a component is referred to as being "connected" to another component, it should be understood that there may be other components in between, although it may be directly connected to the other component. On the other hand, when a component is referred to as being "directly connected" to another component, it should be understood that there is no other component in between. On the other hand, other expressions describing the relationship between the components, such as "between" and "immediately between" or "neighboring to" and "directly neighboring to", should be interpreted as well.

Singular expressions should be understood to include plural expressions unless the context clearly indicates otherwise, and terms such as "comprise" or "have" refer to a feature, number, step, operation, component, part, or feature thereof. It is to be understood that the combination is intended to be present and does not exclude in advance the possibility of the presence or addition of one or more other features or numbers, steps, operations, components, parts or combinations thereof.

In each step, an identification code (e.g., a, b, c, etc.) is used for convenience of description, and the identification code does not describe the order of the steps, and each step clearly indicates a specific order in context. Unless stated otherwise, they may occur out of the order noted. That is, each step may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the reverse order.

The present invention can be embodied as computer readable code on a computer readable recording medium, and the computer readable recording medium includes all kinds of recording devices in which data can be read by a computer system. . Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. Generally, the terms defined in the dictionary used are to be interpreted to coincide with the meanings in the context of the related art, and should not be interpreted as having ideal or excessively formal meanings unless clearly defined in the present application.

1 is a diagram illustrating a spam tag-based blog spam detection system according to an embodiment of the present invention.

Referring to FIG. 1, a spam tag-based blog spam detection system (hereinafter, referred to as a blog spam detection system) 100 may include a user terminal 110, a blog spam detection device 130, and a database 150. Can be.

The user terminal 110 may correspond to a computing device capable of accessing a specific blog to check the contents of the blog, and may be implemented as a smartphone, a notebook, or a computer, and is not limited thereto, and various devices such as a tablet PC. It can also be implemented. The user terminal 110 may be connected to the blog spam detection device 130 through a network, and the plurality of user terminals 110 may be simultaneously connected to the blog spam detection device 130.

In one embodiment, the user terminal 110 may request spam detection for a specific blog from the blog spam detection device 130, and receive and confirm a blog spam detection result from the blog spam detection device 130. In another embodiment, when the user terminal 110 accesses a post page of a specific blog composed of a plurality of posts, the user terminal 110 automatically requests the blog spam detection device 130 for spam detection for the specific blog including the post page. The blog spam detection result may be displayed based on the blog spam detection information received from the blog spam detection device 130.

The blog spam detection device 130 corresponds to a computer or a program capable of detecting spam for a specific blog according to a blog spam detection request received from the user terminal 110 and providing the detection result to the user terminal 110. Can be implemented as a server. The blog spam detection device 130 may be wirelessly connected to the user terminal 110 through Bluetooth, WiFi, or the like, and may exchange data with the user terminal 110 through a network.

Blog spam detection device 130 may be implemented including a database 150, it may be implemented independently of the database (150). When implemented independently of the database 150, the blog spam detection device 130 may be connected to the database 150 by wire or wirelessly to exchange data.

The database 150 is a storage device that can store various information necessary for blog spam detection. The database 150 may store blog information related to a spam detection request received from the user terminal 110, and may store post information in a blog used for blog spam detection and tag information of a corresponding post, but is not limited thereto. Instead, in the process of performing spam detection for a particular blog, information collected or processed in various forms may be stored.

The database 150 may be composed of at least one independent sub-databases that store information belonging to a specific range, and may be configured as an integrated database in which at least one independent sub-databases are integrated into one. When composed of at least one independent sub-database, each sub-database may be wirelessly connected through Bluetooth, WiFi, and the like, and may exchange data with each other through a network. When the database 150 is configured as an integrated database, the database 150 may include a control unit for integrating respective sub-databases into one and managing data exchange and control flow between them.

FIG. 2 is a block diagram illustrating a blog spam detection device of FIG. 1.

2, the blog spam detection apparatus 130 may include a spam tag identification unit 210, a blog reliability calculation unit 230, a blog spam detection unit 250, and a controller 270.

The spam tag identification unit 210 may identify a spam tag among a plurality of tags included in the post based on the body of at least one post posted to the blog. The spam tag identification unit 210 may identify the spam tag based on the relevance of the content of the post body and the tag, and determine the relevance with the tag based on the frequency of the tag appearing in the post body. To identify spam tags. The spam tag identification unit 210 will be described in more detail with reference to FIG. 3.

The blog reliability calculator 230 may calculate the reliability of the blog based on a spam post including all posts and spam tags posted on the blog. Here, the trustworthiness of the blog may correspond to a quantification of the trustworthiness of the blog, and the more the posts posted to the blog include spam tags that are not related to the content of the post body, the more the trustworthiness of the blog Can be measured low.

In one embodiment, the blog reliability calculator 230 may calculate a value obtained by dividing the number of spam posts by the total number of posts as the reliability of the blog. The reliability of a blog may correspond to the spam tag usage index of the blog, and the spam tag usage index may correspond to a ratio of the number of spam posts to the total number of posts posted to the blog. More specifically, the spam tag use index may be higher as the number of posts including spam tags among the posts posted on the blog increases, and the higher the spam tag use index, the lower the reliability of the blog. In other words, the lower the spam tag usage index, the higher the reliability of the blog.

The blog spam detector 250 may detect whether the blog is spam based on the degree of tag spam corresponding to the reliability and the spam tag ratio of the spam post. In one embodiment, the tag spam level may correspond to the ratio of the number of spam tags to the total number of tags. In another embodiment, the tag spam level may correspond to the ratio of the total number of spam tags to the total number of tags of all the posts containing spam tags in one blog. In another embodiment, the tag spam level may correspond to the ratio of the average spam tag number per post to the average tag number per post for the posts containing the spam tag.

The blog spam detection unit 250 may identify a post including the spam tag based on the spam tag identified by the spam tag identification unit 210, and calculate a tag spam degree based on the identified post. have. The blog spam detection unit 250 may determine a tag spam calculation method and detect whether the blog is spam by using the tag spam degree calculated by the method.

In one embodiment, the blog spam detection unit 250 may determine that the blog as spam when the weighted sum between the reliability of the blog and the degree of tag spam exceeds a certain threshold. The blog spam detection unit 250 compares a value obtained by adding the tag spam degree corresponding to the reliability of the blog calculated by the blog reliability calculation unit 230 and the spam tag rate per post with a specific threshold to determine whether the blog is spam. You can judge. Here, the specific threshold value may be preset through the blog spam detection device 130.

For example, the blog spam detection unit 250 may determine whether the blog is spam by comparing a value calculated through the following equation with a specific threshold value.

[Equation]

Here, W ₀ and W ₁ are weights and W ₀ + W ₁ = 1. In addition, the average number of tags per post and the average number of spam tags per post may be calculated for only posts in which spam tags are used.

The higher the trustworthiness of the blog, the higher the spam tag usage index, and the higher the tag spam rate, the higher the spam tag ratio per post, so the higher the sum of the two, the higher the probability that the blog is spam. As a result, the blog spam detection unit 250 may quickly detect whether the blog is spam by determining that the blog is spam when the weighted sum of the reliability and the tag spam degree exceeds a specific threshold.

The controller 270 controls the overall operation of the blog spam detection device 130 and manages a control flow or data flow between the spam tag identification unit 210, the blog reliability calculation unit 230, and the blog spam detection unit 250. can do.

3 is a block diagram illustrating a spam tag identification unit of FIG. 2.

Referring to FIG. 3, the spam tag identification unit 210 may include a post group generation module 310, a tag analysis module 330, a spam tag identification module 350, and a control module 370.

The post group generation module 310 may generate post groups by extracting post key terms through clustering of at least one post and performing clustering based on post key terms. Here, since the post key term is extracted for all texts included in the post body, the post key term may overlap with a post tag included in the post but may not match the post tag.

The tag analysis module 330 may extract group key terms for the post group and perform a comparative analysis between the group key terms and the group tags included in the post group. The tag analysis module 330 may use the post key terms extracted by the post group generation module 310 as they are. The tag analysis module 330 may integrate the core terms of all the posts belonging to the post group into one and use them as key terms for the post group. The tag analysis module 330 may extract various information necessary for tag analysis by comparing the group key terms with a group tag corresponding to a tag included in a post group.

In one embodiment, the tag analysis module 330 may derive the number of occurrences, the number of appearance posts, and the appearance rate in the post group for each group tag through comparative analysis. The frequency of appearance in a post group may correspond to the number of group tags that appear in all posts belonging to a post group. The number of posts in the post group may correspond to the number of posts in which the group tag appears among the posts belonging to the post group. The appearance rate in the post group may correspond to the ratio of the number of posts in which the group tag appears to the number of all posts belonging to the post group.

The spam tag identification module 350 may detect the spam tag using the comparison analysis result. The spam tag identification module 350 may determine whether the spam tag is a spam tag based on statistical information for each group tag, such as the frequency of occurrence, the number of posts, and the rate of occurrence derived through the tag analysis module 330.

In one embodiment, the spam tag identification module 350 may identify a group tag whose spam rate falls below a certain threshold as a spam tag. The spam tag identification module 350 may uniformly determine the spam tag when the percentage of spam tags in the post group derived by the tag analysis module 330 for each group tag is less than a specific threshold, and the specific threshold may be determined as blog spam detection. It may be preset by the device 130.

The control module 370 controls the overall operation of the spam tag identification unit 210 and manages the control flow or data flow between the post group generation module 310, the tag analysis module 330, and the spam tag identification module 350. can do.

4 is a flowchart illustrating a blog spam detection process performed by the blog spam detection device of FIG. 1.

Referring to FIG. 4, the blog spam detection device 130 may use a spam tag among the plurality of tags included in the at least one post based on the body of the at least one post posted to the blog through the spam tag identification unit 210. Can be identified (step S410). The blog spam detection device 130 may calculate the reliability of the blog based on the spam post including the entire post and the spam tag posted on the blog through the blog reliability calculator 230 (step S430). The blog spam detection device 130 may detect whether or not the blog is spam based on the degree of tag spam corresponding to the reliability and the spam tag ratio of the spam post through the blog spam detection unit 250 (step S450).

FIG. 5 is a flowchart illustrating a spam tag identification process performed by the spam tag identification unit of FIG. 2. The spam tag identification unit 210 may extract post key terms through text parsing of at least one post through the post group generation module 310 and generate a post group by performing clustering based on the post key terms. (Step S510). The spam tag identification unit 210 may extract the group key term for the post group through the tag analysis module 330 and perform a comparative analysis between the group key term and the group tag included in the post group (step S530). The spam tag identification unit 210 may detect the spam tag using the comparison analysis result through the spam tag identification module 350 (step S550).

6 is an exemplary view illustrating a process of identifying a spam tag performed by the spam tag identification unit of FIG. 2.

Referring to FIG. 6, the spam tag identification unit 210 may generate a post group through the post group generation module 310, and extract key terms for each post group through the tag analysis module 330. Table 1 610 shows an example of extracting a group tag for each post group PG_ID. Here, the group tag may correspond to a set of tags used in each post belonging to a post group. Freq. (InPost) corresponds to the frequency at which each tag appears in the body of each post per post belonging to a particular post group.

The spam tag identification unit 210 may perform a comparative analysis of key terms and post tags for each post group through the tag analysis module 330. Table 2 (630) shows the frequency of appearance (Freq. (In Post)) in the post body of the group tag per post, the number of posts in which the group tag appeared in the post group to which each group tag belongs (Num. Of Post), Num. Of Post (for each PG) of the corresponding post group and the ratio of the number of posts in the group tag to the total number of posts in the post group to which the group tag belongs (Ratio of Appearance (in PG)). have. The spam tag identification unit 210 may identify the spam tag according to the value of Ratio of Appearance (in PG) through the spam tag identification module 350.

Although described above with reference to a preferred embodiment of the present invention, those skilled in the art will be variously modified and changed within the scope of the invention without departing from the spirit and scope of the invention described in the claims below I can understand that you can.

100: spam tag-based blog spam detection system
110: user terminal 130: blog spam detection device
150: database
210: spam tag identification unit 230: blog reliability calculation unit
250: blog spam detection unit 270: control unit
310: post group generation module 330: tag analysis module
350: spam tag identification module 370: control module
610: table 1 630: table 2

Claims (13)

A spam tag identification unit for identifying a spam tag among a plurality of tags included in the at least one post based on a body of at least one post posted on a blog;
A blog reliability calculator configured to calculate a reliability of the blog based on the entire post posted to the blog and a spam post including the spam tag; And
Spam tag-based blog spam detection device comprising a blog spam detection for detecting whether the blog spam based on the degree of tag spam corresponding to the reliability and the spam tag rate of the spam post.
The method of claim 1, wherein the spam tag identification unit
A post group generation module configured to extract post key terms through text parsing of the at least one post, and to generate a post group by performing clustering based on the post key terms;
A tag analysis module extracting a group key term for the post group and performing a comparative analysis between the group key term and a group tag included in the post group; And
Spam tag-based blog spam detection device comprising a spam tag identification module for identifying a spam tag using the comparison analysis results.
The method of claim 2, wherein the tag analysis module
Spam tag-based blog spam detection device, characterized in that the derivation frequency, the number of appearance posts and the appearance rate in the post group for each group tag through the comparative analysis.
The method of claim 3, wherein the spam tag identification module
Spam tag-based blog spam detection device characterized in that for identifying the group tag that the appearance rate is less than a certain threshold value as a spam tag.
According to claim 1, wherein the blog reliability calculation unit
The spam tag-based blog spam detection device, characterized in that the value obtained by dividing the number of the spam posts by the total number of posts as the reliability of the blog.
According to claim 1, wherein the blog spam detection unit
The spam tag-based blog spam detection device, characterized in that the blog is determined to be spam when the weighted sum between the reliability and the tag spam degree exceeds a specific threshold.
In the blog spam detection method performed in the spam tag-based blog spam detection device,
(a) identifying a spam tag among a plurality of tags included in the at least one post based on a body of at least one post posted to a blog;
(b) calculating a reliability of the blog based on the entire post posted to the blog and a spam post including the spam tag; And
(c) detecting whether the blog is spam based on a tag spam level corresponding to the reliability and the spam tag ratio of the spam post.
The method of claim 7, wherein step (a)
(a1) generating a post group by extracting a post key term through text parsing of the at least one post and performing clustering based on the post key term;
(a2) extracting a group key term for the post group and performing a comparative analysis between the group key term and a group tag included in the post group; And
(a3) spam tag-based blog spam detection method comprising the step of identifying a spam tag using the comparative analysis result.
The method of claim 8, wherein step (a2)
Deriving the spam tag-based blog spam detection method, characterized in that the step of deriving the frequency of occurrence, number of appearance posts and the appearance rate in the post group for each group tag through the comparative analysis.
The method of claim 9, wherein step (a3)
The spam tag-based blog spam detection method, characterized in that the step of identifying the group tag that the appearance rate is less than a certain threshold value as a spam tag.
The method of claim 7, wherein step (b)
And calculating a value obtained by dividing the number of the spam posts by the total number of posts as the reliability of the blog.
The method of claim 7, wherein step (c)
And determining the blog as spam when the weighted sum between the reliability and the tag spam degree exceeds a specific threshold.
Claims [1] A computer-executable recording medium for recording a blog spam detection method performed by a spam tag-based blog spam detection device,
Identifying a spam tag among a plurality of tags included in the at least one post based on a body of at least one post posted to a blog;
Calculating a reliability of the blog based on the entire post posted to the blog and a spam post including the spam tag; And
And detecting whether or not the blog is spam based on the degree of tag spam corresponding to the reliability and the spam tag ratio of the spam post.

Images