JP2009211281A - Propagation structure extraction device, method and program, and computer-readable recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To discriminate the display of articles including keywords designated from a user from the display of articles not including the keywords, and to display the connection of users which can not be found out only from the connection of the articles. <P>SOLUTION: This propagation structure extraction device is configured to extract link information (connection) not only from comments or track-back directly input from an article page but also from a user site including the article page or the article page of a user who has input the comments or track-back or the user site including the article page, and to display a propagation structure with a certain article as a start point or a propagation structure with a blog site as a start point. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a propagation structure extraction apparatus and method, a program, and a computer-readable recording medium. In particular, in a blog or social networking service (SNS) on the Internet, how an article sent by an individual affects a viewer. The present invention relates to a propagation structure extraction apparatus and method, a program, and a computer-readable recording medium that can be grasped by presenting a propagation structure.

  Specifically, a propagation structure extraction device that extracts a propagation structure that an article including a specific keyword has influenced readers by analyzing an article and comments and trackbacks transmitted to other articles in the blog site that includes the article. And a method, a program, and a computer-readable recording medium.

  With the widespread use of blog services and SNSs, anyone can easily send information. The published articles often contain descriptions of daily life and include experiences and evaluations using products and services. By giving a comment to the sent article, the viewer can be notified of the impression of the article. In the blog service, when an article is written with reference to an article, the trackback function can notify the article referred to. Therefore, by analyzing comments and trackbacks given to articles in which word-of-mouth information about products and services is described, it is possible to extract the propagation path and range of the influence of word-of-mouth on viewers.

As a conventional technique, there is a technique for collecting information on a community from comments and trackback information for a blog community (see, for example, Patent Document 1).
JP 2007-80103 A

  However, the above conventional technology sequentially collects articles belonging to the blog of the comment sender and collects articles of the trackback sender, and from the collected data, it propagates from the article and links between the blogs. It does not consider the method of extracting and visualizing. Therefore, it is necessary to extract the propagation structure while sequentially tracing the articles with respect to the influence of the article specified by the user, and there is a problem that it takes time to display the propagation structure.

  In addition, among users who are in close communication with each other, comments and trackbacks may not always be sent to affected articles. For example, user A posted an article about a product called XX such as “I bought XX, but it ’s very comfortable and recommended!” And user B who viewed it posted it on his blog at a later date. You may have written an article such as “I bought a XX that my friend introduced the other day”. In this case, even if there is no comment / trackback from the user B in the article about the user A's XX product, the user A and the user B communicate with each other if a comment or trackback is transmitted to another article. In addition, if it is found that a specific keyword XX is included in both articles, the possibility of being influenced by either article can be discovered.

  FIG. 16A shows an example of the relationship between a blog site, an article, a comment, and a trackback. A comment is information transmitted to an article, and a comment sender has a blog site (a comment from user B to user A's article c), a blog site, or a blog site There is a case of commenting without clearly specifying the URL (comment from user D to article d of user A). The trackback is information transmitted from the article to the article, and an example of the information transmitted from the article a of the user C to the article d of the user A is shown. Comments can be sent without a blog site, but trackbacks occur only between users who have blog sites.

  FIG. 16B shows an example illustrating the propagation structure from the article d of the user A when there is a relationship as shown in FIG. Comment transmission is indicated by a solid line arrow, trackback transmission is indicated by a dotted line arrow, and keywords XX are indicated by satin. Here, attention is paid to the point that the node of the article a and the node of the blog site of the user C are drawn separately. Actually, since the article a is an article included in the blog site of the user C, it is desirable to display it as one node. However, if it is displayed only as the blog site of the user C, the existence of the article C that transmitted the trackback is hidden. Since trackback is very important information as a link between articles, it is desirable to display the article that transmitted the trackback as a node and add comment transmission as additional information. In order to check whether the comment and the trackback sender are the same, it is necessary to store the URL of the blog site in the comment information and the trackback information and check whether the values are the same.

  Also, if the URL of the comment sender is unknown and the name of the comment sender is known, it may be considered as the same person within the same blog site as long as the comment sender name is the same. It is necessary to create a graph.

  Furthermore, the user B usually communicates with the user A, and even if the keyword f is described in the article f, the connection cannot be seen because the comment or the trackback transmission is not performed on the article d. There's a problem.

  The present invention has been made in view of the above points, and can distinguish between the display of an article that includes a keyword specified by the user and the display of an article that is not included, and can be found only by connection from an article. It is an object of the present invention to provide a propagation structure extraction apparatus and method, a program, and a computer-readable recording medium capable of displaying a connection between users who cannot perform the operation.

  FIG. 1 is a principle configuration diagram of the present invention.

The present invention (Claim 1) is a propagation structure extraction device that analyzes the influence of a blog on the Internet or an article of a social networking service on a viewer,
An article identifier for uniquely identifying the article by analyzing the collected article and a blog identifier for uniquely identifying the blog site to which the article belongs are assigned to the article and stored in the article information storage means 11 for comments and trackbacks. Article analysis means 30 for storing the information in the comment / trackback storage means,
A recursive collection unit that identifies a blog site as a transmission source when a comment is transmitted to an article, collects articles belonging to the blog site, and collects an article as a trackback source when a trackback is transmitted 40,
An article link information extracting unit 13-1 that extracts a comment from an article or a link (link information) by a trackback and stores it in the link information storage unit 14;
A blog link information extracting unit 13-2 that collects articles belonging to the same blog site and extracts a link (link information) by a comment or a trackback from the blog site and stores the link information in the link information storage unit 14;
Article propagation display means 15-1 for obtaining and displaying a propagation structure starting from a certain article from the link information storage means 14,
Blog propagation display means 15-2 for obtaining and displaying the propagation structure based on the blog site from the link information storage means 14,
When switching from the article propagation display means 15-1 to the blog propagation display means 15-2, there is provided a switching display means 15-3 starting from a blog site including an article starting from the article propagation display means.

Further, the present invention (Claim 2) is provided in the article analysis means 30.
When a comment is transmitted to an article, an article identifier that uniquely identifies the article, a blog identifier that uniquely identifies a blog site to which the article belongs, at least one of a comment sender name or a comment sender URL, If there is a comment sender URL, it is checked whether it is the URL of the blog site. If it is the URL of the blog site, a blog identifier that uniquely identifies the blog site is stored in the comment / trackback storage means 12 as comment information. Including means.

Further, the present invention (Claim 3) is the article analysis means 30,
When a trackback is transmitted to an article, an article identifier that uniquely identifies the article, a blog identifier that uniquely identifies a blog site to which the article belongs, an article identifier of a trackback transmission source, a blog identifier, an article URL, a blog One or more pieces of information that can be acquired in the URL and the blog title are stored in the comment / trackback storage unit 12 as trackback information.

The present invention (Claim 4) is the article link information extracting means 13-1,
Based on the trackback information stored in the comment / trackback storage means 12, article link information is generated with the trackback received article as the reference article, the trackback transmission article as the connected article, and the comment information set transmitted to the reference article If the blog identifier of the comment sender is included in the generated article link information, the new article link information is not generated and the number of received comments is added to the existing article link information. If not included, the article consists of the article identifier of the article that received the comment, the blog identifier, the blog identifier of the comment sender, the comment sender name and / or the URL of the comment sender, and the number of comments. Means for generating link information;
Comment information with an unknown blog URL is determined to be the same if the blog sender name is the same, and the number of comments is added to the existing article link information. Means for generating;
Means for storing all the generated article link information in the link information storage means 14.

Further, the present invention (claim 5) is the article propagation display means 15-1,
When the starting article and the number of steps following the comment or trackback are input, the article link information in the link information storage means 14 is referred to, and the article link information is sequentially searched until the input number of stages is reached. The article link information includes a means for generating a graph representing propagation from the article and displaying the case where the node includes a specific keyword and the case where the node does not include the specific keyword.

Further, the present invention (Claim 6) is the blog link information extracting means 13-2.
Based on the comment information stored in the comment / trackback storage means 12 and the blog identifier of the trackback information, two blog identifiers having at least one connection between the comment and the trackback, the number of comments transmitted and received, the number of tracksback transmitted and received If the blog identifier is the same as the blog identifier that received the comment and the blog sender name is the same, the existing blog link information A means of adding comments and generating new blog link information if it does not exist,
Means for storing all the generated blog link information in the link information storage means.

Further, the present invention (Claim 7) is provided in the blog propagation display means 15-2.
When the starting blog and the number of steps to follow the comment or trackback are input, the blog link information in the link information storage means is referenced, and the blog link information is sequentially searched until the input number of steps is reached. The link information includes a means for generating a graph having a blog as a node and a connection between blogs as an edge, and displaying the node in a case where the node includes a specific keyword and a case in which the node does not include the specific keyword.

  FIG. 2 is a diagram for explaining the principle of the present invention.

The present invention (Claim 8) is a method for extracting a propagation structure in an apparatus for analyzing the influence of a blog on the Internet or an article of a social networking service on a viewer,
The article analysis means analyzes the collected articles and assigns an article identifier for uniquely identifying the article and a blog identifier for uniquely identifying the blog site to which the article belongs to the article, and stores the article identifier in the article information storage means. (Step 1), an article analysis step (Step 2) for storing comments and trackback information in the comment / trackback storage means;
When a comment is transmitted to an article, the recursive collection means identifies a blog site as a transmission source, collects articles belonging to the blog site, and when a trackback is transmitted, selects the article as the trackback source. A recursive collection step (step 3) to collect;
Article link information extraction means stores an article link information extraction step (step 4) in which a link from the article or a link by link back (link information) is stored in the link information storage means;
An article propagation display step (step 5) in which the article propagation display means acquires and displays the propagation structure starting from a certain article from the link information storage means;
The blog link information extracting means extracts the links (link information) by comments or trackbacks from the blog site by collecting articles belonging to the same blog site, and stores them in the link information storage means (step 6). When,
A blog propagation display step (step 7) in which the blog propagation display means acquires and displays the propagation structure based on the blog site from the link information storage means;
When the switching display means switches from the article propagation display to the blog propagation display, a switching display step (step 8) starting from the blog site including the article starting from the article propagation display step is performed.

  The present invention (Claim 9) is a propagation structure extraction program for causing a computer to function as each means constituting the propagation structure extraction apparatus according to any one of Claims 1 to 7.

  The present invention (Claim 10) is a computer-readable recording medium storing the propagation structure extraction program according to Claim 9.

  As described above, in the present invention, the article link information is extracted from the comment / trackback information transmitted in advance with respect to the article, the blog link information that summarizes the comments / trackback information between the blogs is extracted, and the user specifies the specific information. When an article that includes a keyword is specified, a graph that lists direct links to the article (articles that send trackbacks to the article and connections to senders who send comments) is quickly displayed. It becomes possible.

  Even if you don't comment or track back directly on an article, you can get an overview of propagation by displaying the link information between blogs and the presence of keywords at the same time, and find connections that can't be found by propagation from articles It becomes possible.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 3 shows the configuration of the propagation structure extraction apparatus according to an embodiment of the present invention.

  The propagation structure extraction device includes an article information storage unit 11, a comment / trackback storage unit 12, a link information extraction unit 13, a link information storage unit 14, a propagation display unit 15, a search unit 16, an article collection unit 20, an article analysis unit 30, A recursive collection instruction unit 40 and an interface unit 50 are included.

  Hereinafter, the operation of each component of the propagation structure extracting apparatus will be described with reference to the sequence chart of FIG.

  The article collection unit 20 receives a list of article URLs to be collected, sequentially collects article URLs described in the article URL list (step 101), and outputs the collected HTML text (step 102).

  The article analysis unit 30 analyzes the HTML text received from the article collection unit 20, and assigns an article identifier that uniquely identifies the article and a blog identifier that uniquely identifies the blog site that includes the article to the article. Store in the storage unit 11 (step 103). If a comment is transmitted, the comment information is stored in the comment / trackback storage unit 12. If a trackback is transmitted, the trackback information is stored in the comment / trackback storage unit 12 (step 104). ). The URL of the comment transmission source and the article URL of the trackback source as a result of analysis are sent to the recursive collection instruction unit 40 (step 105).

  When the recursive collection instruction unit 40 receives the URL of the comment transmission source, the recursive collection instruction unit 40 acquires an updated RSS file of the comment transmission source blog (step 106), and sends the article URL list described in the update RSS to the article collection unit 20 ( Step 107). The updated RSS file has a URL unique to the blog site, and is a file that always describes the URL of the latest N articles and an outline thereof. When the article URL of the trackback transmission source is received, the article URL is sent to the article collection unit 20 (step 108).

  The article collection unit 20, the article analysis unit 30, and the recursive collection instruction unit 40 control the articles in the article URL list input by the user to have a recursion count of 0 and perform recursive collection up to a preset number (K times). . That is, the article collection unit 20 sets the recursion count to 0 for the article URL input by the user, and passes it to the article analysis unit 30 together with the collected HTML text. After analyzing the HTML text and storing the comment / trackback information, if the number of recursion is less than K times, the article analysis unit 30 sets the recursion number of the HTML text analyzed to the comment transmission source URL or the trackback source article URL. The number of times obtained by adding 1 is given and sent to the recursive collection instruction unit 40. When the recursion count is K, the comment transmission source URL and the trackback source article URL are not sent to the recursive collection instructing unit 40 so that recursive collection is not performed any more. The recursive collection instructing unit 40 assigns the received recursion count to all URLs acquired from the RSS of the comment transmission source blog, and sends it to the article collecting unit 20. Similarly, the trackback source article URL is sent to the article collection unit 20 with the received recursion count. The article collection unit 20 also performs article URL collection S101 for the URL list received from the recursive collection instruction unit 40.

Also, the article analysis unit 30 generates a list of article identifiers that have been newly analyzed periodically and sends the list to the link information extraction unit 13 (step 109).
The link information extraction unit 13 periodically receives the analyzed article list from the article analysis unit 30, generates article link information indicating connection from articles and blog link information indicating connection between blogs, and stores link information. Store in the unit 14 (step 110).

  The interface unit 50 receives a keyword input from the user (step 111), and passes the keyword to the search unit 16 (step 112). The search unit 16 searches the article information storage unit 11 based on the received keyword (step 113), acquires article information including the keyword in the text or title, and returns it to the interface unit 50 (step 114). The interface unit 50 presents the received article information to the user (step 115), and accepts article designation from the user (step 116). Information about whether the received article identifier and keyword are included is sent to the propagation display unit 15 (step 117). There may be a plurality of keywords given by the user, or the system may perform a search by expanding synonyms and the like based on the keywords inputted by the user. Further, the search is not limited to keyword search (search for information including a keyword), but may be performed using concept search (search for information similar in concept to the keyword) or the like. In this case, information indicating whether the concept is similar or not is transmitted to the propagation display unit 15.

  The propagation display unit 15 receives the article identifier, searches the link information storage unit 14 for nodes connected to the article identifier, and sequentially searches for nodes up to N stages to generate a graph (step 118). Also, the case where each node includes a specific keyword and the case where it does not include a specific keyword are displayed separately (step 119). Further, the user is made to select an article as a new starting point from the displayed propagation structure, and the propagation structure is displayed (step 120). The number of steps to be displayed may be set in the system in advance, or may be configured to be specified by the user.

  Also, an instruction to display the connection between the blogs is received from the user from the propagation structure of the displayed article, and the N-stage blog link information is displayed starting from the blog including the article that is currently the starting point. Also in this case, the possibility of propagation can be found by displaying the blog including the keyword and the blog not including the keyword separately.

  Here, the article page which is the processing target of the present invention will be described. FIG. 5 is a configuration example of an article page targeted by the present invention. Here, a case is shown in which the comment set 202 and the trackback set 203 are described in the article page 200 as text, but they may be stored separately and associated with each other by a hyperlink.

  The article page 200 includes an article body 201, a comment set 202 transmitted to the article body, and a trackback set 203. Each comment in the comment set 202 includes a comment title, a comment sender URL, a comment sender name, a comment transmission date and time, a comment body, and the like. The comment title, the URL of the comment sender, the name of the comment sender, etc. may not be input at the time of transmission. In this case, the system setting value such as “Unknown” is often described.

  Therefore, in the link information extraction unit 13, if the URL of the comment sender is the same, the node is one node. If the URL of the comment sender is not input, the node is the same if the comment sender name is the same. . Further, in the extraction of article link information, when trackback transmission is performed, if the blog URL to which the article of the trackback transmission source belongs and the URL of the comment sender are the same, one node is assumed.

  FIG. 6 shows an example of storing article information in the article information storage unit in one embodiment of the present invention, and FIG. 7 shows an example of storing blog information in the article information storage unit in one embodiment of the present invention.

  As article information stored in the article information storage unit 11, an article identifier 41, a blog identifier 42, an article URL 43, an article posting date 44, an article title 45, an article body 46, a comment reception count 47, a trackback reception count 48, and the like are stored. Is done. As the blog information stored in the article information storage unit 11, a blog identifier 51, a blog site URL 52, a blog site update RSS URL 53, a blog site name 54, a blog owner account 55, and the like are stored.

  The article analysis unit 30 updates the existing information when the URL of the analyzed article already exists in the article information of the article information storage unit 11, and adds the new article identifier to the article information when it does not exist. The information is stored in the information storage unit 11. Further, the blog URL is obtained from the article URL based on the rule, and if the blog URL exists in the blog information, the blog identifier is acquired, and if it does not exist, a new blog identifier is assigned to add the blog information to the article information storage unit 11. To accumulate.

  FIG. 8 shows an example of storage of trackback information in the comment / trackback storage unit according to the embodiment of the present invention. The trackback information includes an article identifier 61 and a blog identifier 62 that have received the trackback, an article identifier 63 of the trackback transmission source, a blog identifier 64, an article title 65, an article URL 66, a blog title 67, a blog URL 68, a transmission date and time 69, and the like. Is done.

  FIG. 9 is a storage example of comment information in the comment / trackback accumulating unit in the embodiment of the present invention.

  The comment information of the comment / trackback storage unit 12 includes the article identifier 71 and the blog identifier 72 of the article that received the comment, the blog identifier 73 of the comment sender, the sender name 74, the URL 75 of the sender, and the information of the comment itself. Are composed of a comment title 76, a comment body 77, a comment transmission date and time 78, and the like. The blog identifier 73 of the blog sender is obtained by searching the blog information on the assumption that the URL of the blog sender is the URL of the blog site. If the blog information does not exist, the blog identifier is stored as none.

  Further, in the case of a comment in which the URL of the comment sender is not described, the blog identifier cannot be searched, so that the comment information without the blog identifier and without the URL of the comment sender is accumulated. The sender name 74 includes “Unknown”.

  Both the comment information and the trackback information are always stored because the article identifier of the article that received the comment and the article identifier of the article that received the trackback are articles to be analyzed.

  Hereinafter, the operation of the link information extraction unit 13 will be described in detail.

  FIG. 10 is a flowchart of the article link extraction process of the link information extraction unit in the embodiment of the present invention. FIG. 11 is a storage example of article link information of the link information storage unit according to the embodiment of the present invention. The article link information of the link information storage unit 14 includes a reference article identifier 801, a blog identifier 802 to which the reference article belongs, The article identifier of the node connected to the reference article (when the node is an article) 803, the blog identifier 804 of the node connected to the reference article, the URL 805 of the reference article, the trackback number 806 received from the node by the reference article, and the reference article Number of trackbacks 807 transmitted to the node, number of comments 808 received by the reference article from the node, number of comments 809 transmitted by the blog including the reference article to the node, node owner name 810, article URL 811 of the node, node blog URL 812, node Article title (if node is an article) 813, node is Consisting of the blog title 814.

  The link information generation unit 13 selects an unprocessed article from the article identification list received from the article analysis unit 30 (steps 201 and 202). If there is an unprocessed article, the trackback information received by the article selected from the comment / trackback storage unit 12 is searched. If trackback information exists, article link information is generated using the selected article as a reference article and the article of the trackback transmission source as a connected node (step 203).

  Next, the comment information received by the reference article is searched to obtain a comment information set (step 204). If there is unprocessed comment information (step 205, Yes), the comment sender's blog identifier is first included in the article link information (FIG. 11: node blog identifier 804) based on the article being processed. Find out. If it is included in the existing article link information (step 206, Yes), new article link information is not generated, and the number of received comments of the existing article link information is incremented by 1 (step 207). If it is not included in the existing article link information (step 206, No), the comment sender name is included in the article link information (FIG. 11: node owner name 810) based on the article being processed. (Step 208). When the comment sender name is in the existing article link information (step 208, Yes), 1 is added to the number of comments received in the existing article link information (step 207). If there is no existing article link information with the same comment sender name, the standard article identifier, the blog identifier, the blog identifier of the comment sender, the comment sender name or the URL of the comment sender, or both, and the comment New article link information consisting of a number is generated (step 209).

  All the generated article link information is stored in the link information storage unit 14 in the form shown in FIG. Further, when the article identifier of the node exists in the generated article link information (that is, the reference article receives the trackback from the node), the reference article a and the article identifier of the node for each of the article link information Search for the presence of article link information (article link information with reference article b and node identifier a) that replaces b (the blog identifier to which the node belongs is B). Stores the number of article link information trackback receptions generated as the number of article link information trackback transmissions. Also, it is searched whether there is article link information with the blog identifier to which the reference article belongs and the article identifier of the node is a. If there is, the article link information generated as the number of comment transmissions of the article link information of the search result is found. The number of received comments is stored (step 210).

  Next, generation processing of blog link information indicating the connection between blogs will be described with reference to FIG.

  FIG. 12 is a flowchart of the blog link information extraction process of the link information extraction unit in the embodiment of the present invention. FIG. 13 is a storage example of blog link information of the link information storage unit according to the embodiment of the present invention. The blog link information of the link information storage unit 14 includes a reference blog identifier 901 and a node connected to the reference blog. Blog identifier 902, trackback number 903 received by the reference blog from the node, trackback number 904 sent by the reference blog to the node, number 905 comments received by the reference blog from the node, number 906 comments sent by the reference blog to the node, node ownership It consists of a person name (comment sender name) 907, a node article URL 908, a node blog URL 909, a node article title (if the node is an article) 910, and a node blog title 911.

  The link information generation unit 13 generates a list of blog identifiers from the list of article identifiers received from the article analysis unit 30 and starts processing.

  First, an unprocessed blog identifier is selected (steps 301 and 302). When there is an unprocessed blog, the trackback information received by the article belonging to the selected blog is searched from the comment / trackback storage unit 12. If trackback information exists, blog link information is generated using the selected blog as a reference and using the blog identifier of the trackback transmission source as a connected node (step 303). At this time, the total number of trackbacks received from the plurality of articles belonging to the blog of the node by the plurality of articles belonging to the reference blog is totaled and stored in the blog link information of the link information storage unit 14.

  Next, comment information received by a plurality of articles belonging to the reference blog is searched from the comment / trackback storage unit 12 to obtain a comment information set (step 304). If there is unprocessed comment information (step 305, Yes), the comment sender's blog identifier is first included in the blog link information (FIG. 13: node blog identifier 902) based on the blog being processed. Find out.

  When it is included in the existing blog link information (step 306, Yes), new blog link information is not generated, and the number of received comments of the existing blog link information is incremented by 1 (step 307). If it is not included in the existing blog link information (step 306, No), the blog sender name is included in the blog link information based on the blog being processed (FIG. 13: node owner name 907). (Step 308). When there is existing blog link information having the same comment sender name (step 308, Yes), 1 is added to the number of comments received in the existing blog link information (step 307). If there is no existing blog link information with the same comment sender name, the reference blog identifier, the blog identifier of the comment sender, the comment sender name is one or both of the URLs of the comment sender, and the number of comments New article link information is generated (step 309).

  All the generated blog link information is stored in the link information storage unit 14 in the form shown in FIG. 13, and for each blog link information in which the blog identifier of the node exists in the generated blog link information, Searches for the existence of blog link information that replaces the node, and if so, stores the number of trackback receptions of the generated blog link information as the number of trackback transmissions of the blog link information of the search results, and stores the blog link information of the search results The number of received comments of the blog link information generated as the number of comment transmission is stored (step 310).

  The link information shown in FIG. 11 and FIG. 13 includes the article titles 813 and 910 of the node, the blog title 814 to which the node belongs, and the blog title 911 of the node, which are indispensable for indicating the propagation structure. The item can be easily identified by displaying the article title and the blog title.

  Hereinafter, the processing of the propagation display unit 15 will be described in detail with reference to FIGS. 11 and 13.

  The propagation display unit 15 receives an article identifier to which information indicating whether or not a keyword is included from the interface unit 50, and searches the link information storage unit 14 for article link information based on the article identifier. Next, when there is an article identifier of a node connected to the reference article, article link information using the node as the reference article is searched. This process is performed sequentially until N stages are reached. Articles with the same article identifier are set as one node, nodes without an article identifier are set as one node when the comment sender name is the same, and the connection of each node is an edge. Generate a graph. At this time, as the display form of the edge, the connection of comment transmission / reception and trackback transmission / reception is combined into one edge, the method of using four edges, the method of combining transmission and reception with two, etc. There is. In addition, the node including the keyword is distinguished and displayed by changing the color or the shape of the node including the keyword and the node not including the keyword. Thereby, it is possible to discover how the influence of the article including the keyword has been propagated.

  FIG. 14 is a display example of a connection structure from articles in the embodiment of the present invention, and is displayed by the propagation display unit 15. In the figure, the node where the article identifier exists is displayed as ◯, the node where the article identifier does not exist is displayed as □, and the node including the keyword is displayed with the color changed to black. In addition, the node a which is the starting point is an example in which the contour is displayed double, so that it can be seen that it is the starting point, and the number of steps is up to two. In connection with an article, a node in which an article identifier exists has a trackback transmission / reception relationship, and indicates that a trackback is transmitted from the node b and the node c to the node a. The arrow from node a to node b indicates that the track back from node a to node b or that the blog owner of node a has sent a comment to node b. In this way, when comment transmission / reception and trackback transmission / reception are displayed with one edge, a function of switching the edge to be displayed by designating only the comment or only the trackback may be prepared.

  In the state where the propagation structure from the article is displayed in this way, when the user designates a specific article and gives an instruction to view the propagation from the article, the N stages starting from the designated article as described above are used. Search for article link information until it reaches, and generate a graph.

  Even if there is no direct connection to the article, the connection between the blogs is displayed as a method of discovering the influence of the article. When the user gives an instruction to switch to connection display between blogs, blog link information based on the blog including the starting article is searched. When there is a node connected to the reference blog, blog link information using the node as the reference blog is searched. This process is performed in sequence until the Nth stage is reached. If the blog identifier is the same, the node is one node. If there is no blog identifier, the comment sender name is the same node. To generate a graph.

  FIG. 15 is a display example of a connection structure between blogs according to an embodiment of the present invention. This figure shows an example in which display switching to the propagation structure between blogs is instructed when the propagation structure from the article is displayed in FIG. As in FIG. 14, a node having a blog identifier is displayed as ◯, a node having no blog identifier is displayed as □, and a node including a keyword is displayed with a color changed to black. In addition, the node “a” that is the starting point displays a double outline so that it can be seen that it is the starting point. It is desirable that the article node and the blog node are displayed in different display forms, but here the same display form is used for ease of explanation. The nodes a to d showing articles in FIG. 14 show blogs in FIG.

  When node a becomes a blog, node g increases, and when node c becomes a blog, nodes h, i, and j increase, and it becomes possible to see more connected nodes. . Also, articles containing keywords in the blogs of nodes b and j are displayed in black. In addition, comments and trackbacks transmitted to the nodes b and g are displayed, and a wider range of propagation structures can be viewed.

  The operation of the components of the propagation structure extraction apparatus shown in FIG. 3 can be constructed as a program, installed in a computer used as the propagation structure extraction apparatus, executed, or distributed via a network. is there.

  Further, the constructed program can be stored in a portable storage medium such as a hard disk, a flexible disk, or a CD-ROM, and can be installed or distributed in a computer.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made within the scope of the claims.

  The present invention can be applied to blog services on the Internet and analysis of articles in SNS.

It is a principle block diagram of this invention. It is a figure for demonstrating the principle of this invention. It is a block diagram of the propagation structure extraction apparatus in one embodiment of this invention. It is a sequence chart of operation | movement of the propagation structure extraction apparatus in one embodiment of this invention. It is a structural example of the article page which this invention makes object. It is a storage example of the article information of the article information accumulation | storage part in one embodiment of this invention. It is a storage example of the blog information of the article information storage part in one embodiment of the present invention. It is a storage example of the trackback information of the comment / trackback storage unit in the embodiment of the present invention. It is a storage example of the comment information of the comment / trackback accumulating unit in the embodiment of the present invention. It is a flowchart of the article link information extraction process of the link information extraction part in one embodiment of the present invention. It is a storage example of the article link information of the link information storage part in one embodiment of this invention. It is a flowchart of the blog search information extraction process of the link information extraction part in one embodiment of the present invention. It is a storage example of the blog link information of the link information storage part in one embodiment of this invention. It is an example of a display of the connection structure from the article in one embodiment of the present invention. It is a display example of the connection structure between blogs in one embodiment of the present invention. It is a figure explaining the structure of the blog service which this invention makes object.

Explanation of symbols

11 article information storage means 12 comment / trackback storage means 13 link information extraction section 13-1 article link information extraction means 13-2 blog link information extraction means 14 link information storage means 15 propagation display section 15-1 article propagation display means 15- 2 Blog propagation display means 15-3 Switching display means 16 Search section 20 Article collection section 30 Article analysis means, article analysis section 40 Recursive collection means, recursive collection instruction section 41 Article identifier 42 Blog identifier 43 Article URL
44 Article registration date 45 Article title 46 Article body 47 Number of received comments 48 Number of received trackbacks 50 Interface section 51 Blog identifier 52 URL of blog site
53 Blog site update RSS URL
54 Blog Site Name 55 Blog Owner Account 61 Article Identifier that Received Trackback 62 Blog Identifier that Received Trackback 63 Article Identifier of Trackback Source 64 Blog Identifier to Which Article of Trackback Source belongs 65 Trackback Source Article Title 66 Trackback Transmission Original article URL
67 Trackback sender blog title 68 Trackback sender blog URL
69 Trackback transmission date and time 71 Article identifier that received the comment 72 Blog identifier that received the comment 73 Blog identifier of the comment sender 74 Comment sender name 75 Comment sender URL
76 Comment title 77 Comment body 78 Comment transmission date and time 200 Article page 201 Article body 202 Comment 203 Trackback 801 Standard article identifier 802 Blog identifier to which the standard article belongs 803 Article identifier of the node connected to the standard article (when the node is an article)
804 Blog identifier 805 of the node connected to the reference article 805 URL of the reference article
806 Number of trackbacks received by reference article from node 807 Number of trackbacks sent by reference article to node 808 Number of comments received by reference article from node 809 Number of comments sent by blog including reference article to node 810 Node owner name 811 Node Article URL
812 Node Blog URL
813 Node article title (if node is an article)
814 The title of the blog to which the node belongs 901 The reference blog identifier 902 The blog identifier of the node connected to the reference blog 903 The trackback number 904 that the reference blog received from the node 904 The trackback number that the reference blog sent to the node 905 The reference blog received from the node Number of comments 906 Number of comments sent to the node by the reference blog 907 Owner of the node (comment sender name)
908 Article URL of the node
URL of the 909 node blog
910 Node article title (if the node is an article)
911 Node Blog Title

Claims (10)

Propagation structure extraction device that analyzes the impact of Internet blogs and social networking service articles on viewers,
An article identifier for uniquely identifying the article by analyzing the collected article and a blog identifier for uniquely identifying the blog site to which the article belongs are assigned to the article and stored in the article information storage means for comment and trackback Article analysis means for storing the information in the comment / trackback storage means,
A recursive collection that identifies a blog site as a transmission source when a comment is transmitted to the article, collects articles belonging to the blog site, and collects an article as the trackback source when a trackback is transmitted Means,
Article link information extracting means for storing the link from the article or the link by the trackback (link information) in the link information storage means;
A blog link information extracting unit that collects articles belonging to the same blog site, extracts a link from the blog site or a link by linkback (link information), and stores the link information in the link information storage unit;
Article propagation display means for acquiring and displaying a propagation structure starting from a certain article from the link information storage means;
A blog propagation display means for acquiring and displaying a propagation structure based on a blog site from the link information storage means;
When switching from the article propagation display means to the blog propagation display means, a switching display means starting from a blog site that includes an article starting from the article propagation display means;
A propagation structure extracting apparatus characterized by comprising: The article analysis means includes
When a comment is transmitted to the article, an article identifier that uniquely identifies the article, a blog identifier that uniquely identifies the blog site to which the article belongs, and at least one of a comment sender name and a comment sender URL If there is a comment sender URL, the URL of the blog site is checked. If it is the URL of the blog site, a blog identifier that uniquely identifies the blog site is stored as comment information in the comment / trackback storage means. The propagation structure extracting apparatus according to claim 1, further comprising: The article analysis means includes
When the trackback is transmitted to the article, the article identifier that uniquely identifies the article, the blog identifier that uniquely identifies the blog site to which the article belongs, the article identifier of the trackback transmission source, the blog identifier, the article URL, The propagation structure extracting apparatus according to claim 1, further comprising means for storing, in the comment / trackback storage means, one or more pieces of information that can be acquired in a blog URL and a blog title as trackback information. The article link information extracting means includes:
Based on the trackback information stored in the comment / trackback storage means, the link information is generated as the link back received article as the reference article, the track back transmission article as the article having connection, and the comment information set transmitted to the reference article If the blog identifier of the comment sender is included in the generated article link information, the new article link information is not generated and the number of received comments is added to the existing article link information. If not included, the article consists of the article identifier of the article that received the comment, the blog identifier, the blog identifier of the comment sender, the comment sender name and / or the URL of the comment sender, and the number of comments. Means for generating link information;
Comment information with an unknown blog URL is determined to be the same if the blog sender name is the same, and the number of comments is added to the existing article link information. Means for generating;
Means for storing all the generated article link information in the link information storage means;
The propagation structure extraction device according to claim 1, comprising: The article propagation display means is:
When the article to be the starting point and the number of steps to follow the comment or trackback are input, the article link information in the link information storage means is referenced, and the article link information is sequentially searched until the input number of stages is reached. 5. The propagation structure extracting apparatus according to claim 1, further comprising: means for generating a graph representing propagation from an article in the article link information of the article, and displaying a case where a node includes a specific keyword and a case where the node does not include the specific keyword. The blog link information extracting means includes
Based on the comment information stored in the comment / trackback storage means and the blog identifier of the trackback information, two blog identifiers having at least one connection of a comment or trackback, the number of comments transmitted and received, the number of trackback transmissions If the blog identifier of the comment sender is unknown and the blog identifier is the same as the received blog identifier and the same blog sender name exists, the existing blog link is generated. A means to add the number of comments to the information and to generate new blog link information if it does not exist,
Means for storing all the generated blog link information in the link information storage means;
The propagation structure extraction device according to claim 1, comprising: The blog propagation display means
When the starting blog and the number of steps following the comment or trackback are input, the blog link information in the link information storage means is referenced, and the blog link information is sequentially searched until the input number of steps is reached. The blog link information includes a means for generating a graph having a blog as a node and a connection between blogs as an edge, and displaying the node separately when the node includes a specific keyword and when the node does not include the specific keyword. Or the propagation structure extraction device according to 6. A method for extracting propagation structure in a device that analyzes the influence of articles on blogs and social networking services on the Internet.
Article analysis means analyzes the collected articles and assigns an article identifier that uniquely identifies the article and a blog identifier that uniquely identifies the blog site to which the article belongs to the article and stores it in the article information storage means And an article analysis step for storing comments and trackback information in the comment / trackback storage means;
When the comment is transmitted to the article, the recursive collection means identifies the blog site of the transmission source, collects the articles belonging to the blog site, and when the trackback is transmitted, the article of the trackback source A recursive collection step to collect
Article link information extracting means stores an article link information extracting step for storing a link or link by link back (link information) from the article in the link information storage means;
An article propagation display means for acquiring and displaying a propagation structure starting from a certain article from the link information storage means; and
The blog link information extracting means extracts the links (link information) by comments or trackbacks from the blog site by collecting articles belonging to the same blog site, and stores them in the link information storage means.
A blog propagation display means for acquiring and displaying a propagation structure based on a blog site from the link information storage means;
When the switching display means switches from the article propagation display to the blog propagation display, a switching display step starting from the blog site including the article starting from the article propagation displaying step;
A method for extracting a propagation structure characterized by   The propagation structure extraction program for functioning a computer as each means which comprises the propagation structure extraction apparatus of any one of Claims 1 thru | or 7.   A computer-readable recording medium storing the propagation structure extraction program according to claim 9.

Images