JP5092252B2 - Trend analysis server and trend analysis method - Google Patents

Description

  The present invention relates to a server and a method for analyzing information published on the Internet and providing information related to a trend.

  On the Internet, a large number of people publish various information on homepages and blogs, and the Internet is now a useful source of information that can be obtained in real time. In recent years, attempts have been made to analyze trends using information sources transmitted by individuals who can be called consumers.

  One technique for analyzing trends using Internet information sources is to analyze current trends from search keywords input to a search engine or the like. This method totals search keywords every certain period (for example, one week), and provides information on the current trend from the total results (for example, ranking of search keywords).

  The device disclosed in Patent Document 1 is a device that analyzes a trend using a search keyword, investigates the degree of association between search keywords included in the top ranking of the search keyword, and the user selects one of the rankings. When one search keyword is selected, the search keyword is a device that transmits information such as other search keywords and related information (for example, news articles) that are highly related to the search keyword to the user.

  However, in the method using search keywords, trend analysis is performed based only on keywords that are explicitly specified by the user, so keywords that exist on the Web page but are not specified for search are excluded. , There is a problem of lack of completeness as a trend analysis.

  In addition, as another method of analyzing a trend by using an information source on the Internet, there is a method of collecting information published on the Internet by crawling and analyzing the trend from the collected information.

The device disclosed in Patent Document 2 is a device that applies this method, and automatically collects documents from a plurality of homepages on the Internet and analyzes words included in the collected documents. A device that displays at a glance what is being talked about (= trend). Preferably, the document to be collected is news, so that the frequency of occurrence of important words in the news and the It is a device that allows you to understand related words at a glance.
JP 2004-348554 A JP 2002-108937 A

  However, the technology described above is a technology that provides a word (trend keyword) that indicates the current trend, and understands why the trend keyword is frequently used and information about the information sender that sends the trend keyword. It was difficult to do.

  For example, when the trend keyword was the title of a novel, it was difficult to know whether the novel itself was drawing attention or whether the novel drama was drawing attention. Furthermore, it was difficult to know who was interested in this novel.

  Therefore, the present invention provides a trend keyword that changes with the passage of time to the user, and a trend analysis server that can also provide information on the reason why the trend keyword is used and the person who uses the trend keyword. And it aims at providing a trend analysis method.

A first invention that solves the above-described problem is a trend analysis server that analyzes information published on a plurality of blog sites on the Internet and provides an analysis result to a terminal device connected to the Internet, The trend analysis server obtains, for each blog article published on the blog site, information on the blog site on which the blog article is published and the title, content, and publication date of the blog article as blog article information. Analyzing all the blog article information that meets information collection means and conditions (target article conditions) that define the blog article information to be referred to when extracting keywords, and using one or more words as keywords according to a predetermined criterion Keyword extracting means for extracting, and keywords extracted by the keyword extracting means Web page generating means for generating a Web page to be displayed, wherein the Web page generating means includes a keyword obtained by extracting the blog article information of the blog article published within a first period as the target article condition. The selected trend keyword is described when one trend keyword is selected from the first Web page to be displayed as the trend keyword and the trend keyword displayed on the first Web page. Among the blog article information, a keyword (related keyword) extracted from the latest predetermined number of the blog article information as the target article condition, and the blog article information referred to when the related keyword is extracted , list describes the name of the blog site that is included in the blog post information And, a second Web page that displays, when the name of the blog sites that are displayed has been selected, in the blog post information obtained from the selected the blog site, the latest of a predetermined number The blog article information is used as the target article condition, and a keyword (blog site keyword) extracted using the keyword extracting unit and a list of the blog article information referred to when the blog site keyword is extracted are displayed. A third Web page is generated.

  Further, the second invention is the trend analysis server according to the first invention, comprising an information storage means for storing the blog article information collected by the blog article information collecting means, and the keyword extracting means, It is a means for extracting a keyword using the blog article information stored in the information storage means.

  Further, a third invention is the trend analysis server according to the first invention or the second invention, wherein the keyword of the keyword extracted by the keyword extracting unit is included in the Web page generated by the Web page generating unit. The cloud is displayed.

  Further, a fourth invention is the trend analysis server according to any one of the first to third inventions, wherein the blog article information that the trend analysis server acquires from a plurality of the blog sites is the It is characterized by being all or part of the meta information published on the blog site.

Further, a fifth invention is the trend analysis server according to any one of the first invention to the fourth invention, wherein when listing the blog article information referred to when the keyword is extracted, said in addition to the name of the blog site, the blog said blog post that is included in the article information title, it called, is characterized in that to write a summary statement of the blog post obtained from the blog post information.

  Further, a sixth invention is the trend analysis server according to any one of the first to fifth inventions, wherein the blog article information referred to when the keyword is extracted is listed and displayed. As described above, index information indicating a result of evaluating the blog site from which the blog article information has been acquired with a predetermined index is included.

  Further, a seventh invention is the trend analysis server according to the sixth invention, wherein as the index for evaluating the blog site, the number of linked links of the blog site, the update frequency of the blog site, and the All or part of the number of accesses to the blog site via a trend analysis server is used.

Further, an eighth invention is the trend analysis server according to any one of the first invention to the seventh invention, wherein the trend analysis server is any one of claims 1 to 7, The Web page generation means includes a graph showing the trend frequency of the trend keyword and the transition of the ranking information in the second Web page.

Furthermore, a ninth invention is the trend analysis server according to any one of the first to eighth inventions , wherein the third Web page generated by the Web page generating means includes the third Web page. The blog article information list displayed on the Web page does not include the index information of the site, and the index information of the blog site is displayed separately from the blog article information list. Features.

Further, a tenth invention is the trend analysis server according to any one of the first invention to the ninth invention, wherein the web page generation means generates the third web page when the third web page is generated. The blog article information in which all of the blog site keywords to be displayed on the third Web page are described is extracted, and the latest predetermined number of the extracted blog article information is listed on the third Web page. It is characterized by displaying.

Further, an eleventh invention is the trend analysis server according to any one of the fifth to seventh inventions, wherein the web page generation means is the blog displayed in the blog article information list. When an article title is selected, the blog article information from which the blog article title is selected is used as the target article condition, and a keyword (entry keyword) extracted using the keyword extracting unit, and the entry keyword A fourth Web page that generates a list and displays only the blog article information referred to when extracted is generated.

Furthermore, a twelfth invention is the trend analysis server according to any one of the first invention to the eleventh invention, wherein the Web page generation means is the second period shorter than the first period. The blog article information of the blog article published within is set as the target article condition, the keyword (new keyword) extracted using the keyword extracting means, and the blog article information referred to when the new keyword is extracted And generating a fifth Web page that displays the list.

Furthermore, a thirteenth invention is a trend analysis method for analyzing information published on a plurality of blog sites on the Internet and providing an analysis result to a terminal device connected to the Internet, wherein the trend analysis method includes: For each blog article published on the blog site, blog article information that collects information on the blog site on which the blog article is published, the content of the blog article, and the publication date of the blog article as blog article information Analyzing all the blog article information that meets the conditions (target article conditions) that define the blog article information to be referred to when collecting keywords and extracting keywords, and extracting one or more words as keywords according to predetermined criteria And a Web page generation process for generating a Web page that displays at least the extracted keywords. In the Web page generation step, a first Web page that displays a keyword obtained by extracting the blog article information of the blog article published within a first period as the target article condition as a trend keyword, and the first When one trend keyword is selected from among the trend keywords displayed on the Web page, the latest predetermined number of the blog articles in the blog article information in which the selected trend keyword is described The keywords (related keywords) extracted from the information as the target article condition and the blog article information referred to when the related keywords are extracted are listed by including the name of the blog site included in the blog article information. , and the second of the Web page that displays the name of the blog sites that are displayed When the is selected, among the blog article information obtained from the selected blog site, a keyword (blog site keyword) that extracted the latest predetermined number of the blog article information as the target article condition, A third Web page displaying the list of blog article information referred to when the blog site keyword is extracted is generated.

Furthermore, the fourteenth invention is the trend analysis method according to the thirteenth invention, wherein the blog article information collecting step is a step of storing the collected blog article information in an information storage device, and the web page In the generating step, keywords are extracted using the blog article information stored in the information storage device in the blog article information collecting step.

The fifteenth invention is the trend analysis method according to the thirteenth or fourteenth invention, wherein a keyword cloud of the extracted keywords is displayed on the web page generated in the web page generation step. It is characterized by being.

Furthermore, the sixteenth invention is the trend analysis method according to any one of the thirteenth to fifteenth inventions, wherein the trend analysis server acquires the blog article information acquired from a plurality of the blog sites, It is characterized by being all or part of the meta information published on the blog site.

The seventeenth invention is the trend analysis method according to any one of the thirteenth to sixteenth inventions, wherein the blog article information referred to when the keyword is extracted is displayed in a list. to the addition to the name of the blog site, the blog post information the posts included in the title, your and characterized by the inclusion of summary derived from the posts.

Furthermore, an eighteenth invention is the trend analysis method according to any one of the thirteenth to seventeenth inventions, wherein the blog article information referred to when the keyword is extracted is listed and displayed. As described above, index information indicating a result of evaluating the blog site from which the blog article information has been acquired with a predetermined index is included.

Further, a nineteenth invention is the trend analysis method according to the eighteenth invention, wherein the number of linked links of the blog site, the update frequency of the blog site, and the index as the index for evaluating the blog site, All or part of the number of accesses to the blog site via a trend analysis server is used.

Furthermore, a twentieth invention is the trend analysis method according to any one of the thirteenth to nineteenth inventions, wherein the Web page generation step includes the appearance frequency of the trend keyword and the transition of the ranking information. Is included in the second Web page.

Further, the invention of twenty-first, from the thirteenth aspect a trend analysis method according to any one of the twentieth invention, in the third Web pages generated by the Web page generation step, the first The index information of the blog site is displayed separately from the list of blog article information without including the index information of the site in the list of blog article information displayed on the Web page 3. Features.

Further, a twenty-second invention is the trend analysis method according to any one of the thirteenth to twenty-first inventions, wherein, in the Web page generation step, when the third Web page is generated, The blog article information in which all of the blog site keywords to be displayed on the third Web page are described is extracted, and the latest predetermined number of the extracted blog article information is listed on the third Web page. It is characterized by displaying.

Furthermore, a twenty- third invention is the trend analysis method according to any one of the seventeenth to nineteenth inventions, wherein in the Web page generation step, the blog displayed as a list of the blog article information When an article title is selected, a keyword (entry keyword) extracted from the blog article information from which the blog article title is selected as the target article condition, and the blog referred to when the entry keyword is extracted A fourth Web page that lists and displays only article information is generated.

Furthermore, a twenty-fourth invention is the trend analysis method according to any one of the thirteenth to twenty-third inventions, wherein, in the Web page generation step, the second period shorter than the first period. A keyword (new keyword) extracted as the target article condition for the blog article information of the blog article published in the list, and a list of the blog article information referred to when the new keyword is extracted is displayed. A Web page is generated.

According to the present invention described above, the blog article information of the blog article published on the blog site on the Internet is analyzed to extract the trend keyword, and the relation obtained from the blog article information in which the trend keyword is described. By presenting the keyword to the user, the user can infer from the related keyword the reason why the trend keyword is frequently used. Also, by generating the third Web page that displays the blog site keyword, the user can know the genre mainly described in the blog site.

  Furthermore, by displaying keywords in the keyword cloud, the user can identify important keywords at a glance. Further, by collecting blog article information in advance, it is possible to increase the efficiency of the keyword extraction process, and by using the blog article information as meta information, the efficiency of the keyword extraction process is further increased.

Furthermore, by displaying index information indicating the result of evaluating the site of the blog article with a predetermined index on the Web page, the user can know information such as the popularity of the blog site. Further, by displaying a graph showing the trend frequency of the trend keyword and the transition of the ranking information, the user can know when the trend keyword has started to be used frequently.

In addition, before Symbol to extract the third the blog post information that all of the blog site keywords are described to be displayed on the Web page of, it has been extracted said by displaying a list of the blog post information, the user of the third You can find blog sites that contain articles similar to the blog sites introduced on the Web page.

  Furthermore, by generating the fourth Web page that displays the entry keyword, the user can guess the content of the article corresponding to the blog article information without reading the article body.

  Furthermore, by generating a fifth Web page that displays new keywords, the user can know the keywords that are frequently used when accessing the trend analysis server.

[Trend analysis server]
From here, the trend analysis server which is an apparatus according to the present invention will be described in detail with reference to the drawings. FIG. 1 is a diagram showing an example of a network system in which a trend analysis server is arranged.

  In the network system shown in FIG. 1, a blog management server 2 that provides a blog service and has a plurality of blog sites, an update notification ping server 3 to which update notification ping information indicating that the blog site has been updated is transmitted, The trend analysis server 1 that collects blog article information from the blog site and provides the analysis result of the collected blog article information to the PC 4, and the personal computer 4 possessed by the user using the trend analysis server 1 (hereinafter referred to as PC: Personal) Computer) is connected to the Internet 5.

  In FIG. 1, the blog management server 2 and the update notification ping server 3 are provided as one unit in order to explain the contents of the present invention in an easy-to-understand manner. As a matter of course, a plurality of blog management servers 2 and update notification ping servers 3 are actually connected to the Internet 5. Furthermore, although the trend analysis server 1 is illustrated as being composed of one server, it is actually composed of a plurality of servers.

  The trend analysis server 1 is an application server that collects blog article information from a plurality of blog sites of the blog management server 2 and provides the user with an analysis result of the collected blog article information when requested by the user. In order to collect blog article information from the blog site, the trend analysis server 1 accesses the update notification ping server 3 at predetermined intervals, and acquires blog update information in which a plurality of update communication ping information is aggregated.

  Then, from the blog site indicated by the blog update information, the meta information describing the summary sentence of the updated article and the information (for example, URL) related to the Web page to which the updated article is linked are updated. Collected as blog article information, and the collected blog article information is stored.

  When the trend analysis server 1 is accessed from the PC 4, the trend analysis server 1 blog articles that meet the conditions (target article conditions) that determine the blog article information to be used when extracting keywords from the collected blog article information A keyword is extracted from the information, and a Web page displaying a keyword cloud of the extracted keyword is generated and distributed to the PC 4.

  By changing the target article condition to be used when extracting keywords for each Web page, the group of blog article information referred to when extracting keywords is different, and the extracted keywords have different meanings for each Web page have.

  For example, if words that are frequently used by various blog sites are extracted as trend keywords, and the blog article information in which a particular trend keyword is described is the target article condition, the blog article in which the trend keyword is described Words frequently used in information are extracted as keywords, and the reason why trend keywords are frequently used can be estimated from the keywords.

  Also, if the blog article information collected from a specific blog site is used as the target article condition, the words frequently used by the blog site are extracted as keywords, and the genre mainly described in the blog site can be inferred from this keyword .

  Here, the keyword cloud is a method for displaying a list of keywords. In this keyword cloud, the size, thickness, color scheme, etc. of characters are used according to the appearance frequency and importance. It is displayed more prominently than the keyword. The user can recognize important keywords at a glance by referring to the keyword cloud.

  FIG. 2 is a block diagram of the network system shown in FIG. In the blog management server 2, a large number of people operate the blog site 20 using the blog service provided by the blog management server 2, and daily events and the like are described in one blog site 20. A plurality of blog articles 21 and a piece of meta information 22 which is information describing a summary sentence of the blog article 21 are disclosed.

  FIG. 3 is a diagram for explaining an example of the meta information 22 published on the blog site 20. The meta information 22 includes at least a header element indicating a version of a structured language in which the meta information 22 is described, one Channel element 220, and a plurality of Item elements 221. There are various formats for describing the meta information 22, such as RSS and ATOM. FIG. 3 shows RSS as an example.

  The channel element 220 of the meta information 22 is information that introduces the blog site 20. The title of the blog site 20, the URL at which the blog site 20 is published, and the name of the blog site 20 that publishes the blog site 20. Information such as the theme of the blog site 20 and a list of blog articles 21 are described.

  The Item element 221 is information that is described in the meta information 22 for each blog article 21 of the blog site 20 and describes the summary of the blog article 21. The Item element 221 includes the title of the blog article 21, and the blog article. The URL at which 21 is published, the date and time when the blog article 21 was published, the summary sentence of the blog article 21, and the like are described.

When the blog creator updates the blog site 20, that is, when a new blog article 21 is added to the blog site 20, the meta information 22 is also updated by adding an Item element 221 of the added blog article 21. Further, update notification ping information indicating that the blog site 20 has been updated by the blog creator and indicating the URL of the added blog article 21 is transmitted to the update notification ping server 3.

  The update notification ping server 3 stores update notification ping information transmitted from various blog sites 20, and distributes blog update information 30 in which the transmitted update notification ping information is aggregated. The trend analysis server 1 can acquire the blog update information 30 of the blog site 20 updated within a certain time by accessing the update notification ping server 3 every certain time.

The trend analysis server 1, using a blog update information 30 obtained from the update notification ping server 3, the blog post information collection means 12 for collecting the blog post information, the blog post information that blog post information collection means 12 has been collected Information storage means 13 to be stored, keyword extraction means 10 for analyzing the blog article information stored in the information storage means 13 and extracting keywords, and using the keyword extraction means 10 in accordance with a request from the PC 4 Web page generation means 11 for extracting a keyword, dynamically generating a Web page for displaying the extracted keyword, and distributing it to the PC 4 is provided.

  In the present embodiment, the blog article information collection means 12 of the trend analysis server 1 accesses the update notification ping server 3 at regular intervals and acquires the blog update information published on the update notification ping server 3.

  When the blog post information collection unit 12 of the trend analysis server 1 acquires the blog update information 30, it accesses the URL indicated by the blog update information 30 and is included in the meta information 22 as the blog post information of the added blog post 21. The Channel element 220 and the Item element 221 of the added blog article 21, and information (hereinafter referred to as link information) related to the link destination to which the added blog article 21 is linked are collected. It is stored in the information storage means 13 as article information.

  Blog article information is stored in the information storage means 13 in a database structure, and this database has various data tables. For example, a data table for storing blog article information in time series or a data table generated for each blog site 20. When the blog article information collecting means 12 collects the blog article information, it updates the contents of these data tables based on the collected blog article information.

  In the present embodiment, the Item element 221 of the meta information 22 describing the summary of the added blog article 21 is collected as information describing the content of the added blog article 21, but the trend analysis server 1 May collect the main body of the blog article 21 instead of the Item element 221 of the meta information 22. Even if the main body of the blog article 21 is collected, the technical idea of the present invention does not change as long as the analysis target is changed from the Item element 221 of the meta information 22 to the main body of the blog article 21.

  The keyword extraction means 10 of the trend analysis server 1 stores a plurality of conditions (target article conditions) that determine blog article information used when extracting keywords, and is specified from blog article information that matches the specified target article conditions. This is a means for extracting as many words as keywords.

  The blog article information used when the keyword extracting means 10 for extracting keywords extracts keywords is stored in the information storage means 13, and the target article conditions for extracting keywords differ depending on the Web page to be generated. By extracting keywords from blog article information that matches different target article conditions, the extracted keywords have different meanings for each target article condition, as described above.

  When the keyword extraction means 10 extracts keywords, individual sentences of blog article information (for example, Item elements of meta information) are decomposed into parts of speech by Japanese morphological analysis, which is a well-known technique, and nouns and proper nouns are obtained. And cut out. For the extracted words (nouns, proper nouns), the appearance frequency is calculated for each word and sorted in descending order of appearance frequency. Then, some words (for example, 5 words) at the top of the sorting result are extracted as keywords.

  In general, frequently occurring phrases are often enclosed in parenthesis symbols. Therefore, keywords can be extracted by extracting words enclosed in parentheses instead of morphological analysis. Furthermore, the keyword extraction means 10 may include a plurality of keyword extraction methods, and the extraction method may be changed depending on the keywords to be extracted. For example, when extracting trend keywords, words enclosed in parentheses may be extracted, and other keywords may be extracted using natural language analysis.

  The web page generation unit 11 of the trend analysis server 1 is a unit that dynamically generates a web page using the blog article information stored in the information storage unit 13 and distributes the generated web page to the PC 4. Since the contents of the blog site 20 published on the blog management server 2 are updated every moment, the contents of these Web pages generated by the Web page generation means 11 of the trend analysis server 1 are also accessed by the user It depends on.

  FIG. 4 is a diagram for explaining a Web page generated by the Web page generation unit 11. In the present embodiment, as shown in FIG. 4, a trend keyword page 100 on which a plurality of trend keywords are displayed and a blog in which one selected trend keyword is described as a Web page on which the extracted keywords are displayed. Related article introduction page 110 introducing article 21, blog site introduction page 120 introducing blog site 20 posted on related article introduction page 100, and entry of one blog article 21 posted on related article introduction page 120 At least five Web pages are generated: an entry page 130 that displays information and a new blog page 140 that displays information about newly arrived blog articles.

  First, the trend keyword page 100 that is the top page of the blog analysis site will be described. FIG. 5 is a diagram for explaining the trend keyword page 100.

  On the trend keyword page 100, the blog article information of the blog article 21 updated on the access date to the trend analysis server 1 is set as the target article condition, and the keyword extracted using the keyword extracting means 10 is displayed as the trend keyword. The top 5 trend keywords and the keyword cloud 101 for trend keywords are displayed.

  In FIG. 5, only seven trend keywords are described in the keyword cloud 101, but the number of trend keywords displayed on the keyword cloud 101 may be arbitrary. By referring to the trend keyword keyword cloud 101 displayed on the trend keyword page 100, the user can know what words are frequently used in the vicinity of the access date.

  Further, the trend keyword page 100 is extracted using the keyword extraction means 10 with the blog article information (Item element) of the blog article 21 updated last week on the date of access to the trend analysis server 1 as the target article condition. The top 5 trend keywords from last week are also displayed as reference information.

When any one of the trend keywords displayed on the trend keyword page 100 is selected, the web page generation unit 11 generates a related article introduction page 110 for the selected trend keyword. FIG. 6 is a diagram for explaining the related article introduction page 110 .

  The related article introduction page 110 includes keywords extracted from the blog article information included in the related article introduction list 115 in addition to the related article introduction list 115 that is a list of blog article information in which the selected trend keyword is described. A keyword cloud 111 of a related keyword is displayed.

  By referring to the keyword cloud 111 of related keywords, the user can know the words that appear frequently in the blog articles 21 in which the trend keywords are used, and for what reason the trend keywords are attracting attention. Understandable.

  FIG. 6 is a blog introduction list when “Sunday” is selected on the trend keyword page 100 of FIG. 5. In the related keyword keyword cloud 111, “movie” and “TV” are highlighted. In other words, it can be assumed that “Sunday” is attracting attention in a movie aired on TV on Sunday.

  In the related article introduction page 110 in FIG. 6, the latest five blog article information in which the selected trend keyword is described are included in the related article introduction list 115, and target article conditions for extracting the related keyword are set. Blog article information (Item element) to be included in the related article introduction list 115 is used.

  Furthermore, it is desirable that the number of blog article information items included in the related article introduction list 115 on the related article introduction page 110 in FIG. 6 (five in FIG. 6) can be changed. For example, in the present embodiment, the number of blog articles 21 to be included in the related article introduction list 115 can be increased to 30 by clicking on a location described as “MORE” in FIG. When the number of blog article information is increased, all the displayed blog article information (Item element) becomes a target article condition when a related keyword is extracted.

Further, in the related article introduction list 115 displayed on the related article introduction page in FIG. 6, for each blog article information, the title of the blog article 21 obtained from the blog article information (Item element), the name of the blog site 20, and the blog article. 21 summary text and index information 114 indicating the result of evaluating the blog site 20 from which the blog article information has been acquired are displayed. Further, separately from the related article introduction list 115 , the transition graph of the ranking ranking of the selected trend keyword 112 is displayed. A link to the entry page 130 is attached to the title of the blog article 21 in the related article introduction list 115, and a link to the blog site introduction page 120 is attached to the name of the blog site 20.

  The reason why the index information 114 of the blog site 20 is displayed is to give the user reference information of the blog site 20 when referring to the blog article 21. FIG. 7 is a diagram for explaining the index information 114 of the blog site 20. In this embodiment, the blog article linked to the blog article 21 published on the blog site 20 as the index information 114 of the blog site 20. The number of linked information indicating the number 21, the update frequency information indicating the frequency of updating the blog site 20, and the access count information indicating the number of times the blog site 20 is accessed via the trend analysis server 1 are displayed.

  In FIG. 7, regarding the number of linked information, five levels are set in the number of linked information, and this level is displayed using a star-shaped figure. Regarding the update frequency information, five levels are set in the update frequency information, and this level is displayed using a square. Regarding the access count information, the access count information displays the access count itself as a number.

  The number of linked information is obtained by analyzing the link information collected at the same time when the blog article information is collected, and counting the number of links to the blog site 20. The update frequency information is obtained by calculating the average value of the update intervals of the blog site 20 by analyzing the Item elements of the blog article information collected from the blog site 20. Further, the access count information can be obtained by the blog analysis server storing the number of accesses to the blog site 20 via the trend analysis server 1.

  By referring to the index information 114 of the blog site 20 described above, the user can grasp the degree of popularity of the blog site 20 and can be used as reference information when browsing the blog article 21.

  FIG. 8 is a diagram for explaining the trend keyword ranking transition graph 112 displayed on the related article introduction page 110. The vertical axis of the transition graph 112 is the number of appearances of the trend keyword and the ranking ranking of the trend keyword, and the horizontal axis is the time axis. When the trend analysis server 1 sorts the number of appearances of words in descending order and extracts trend keywords, the trend analysis server 1 stores the number of appearances and the ranking ranking of the sorting results for each trend keyword, and generates this transition graph 112.

  The reason why the transition graph 112 of the ranking ranking of the selected trend keyword is displayed is to provide the user with temporal information regarding the ranking ranking of the selected trend keyword. By referring to the transition graph 112, the user can grasp when the trend keyword has started to be noticed.

  When one of the names of the blog sites 20 displayed in the related article introduction list 115 displayed on the related article introduction page 110 is selected, a blog site introduction page introducing the selected blog site 20 is displayed. FIG. 9 is a diagram for explaining the blog site introduction page 120.

  The blog site introduction page 120 includes a blog site introduction information 122 that introduces the selected blog site 20, a site article introduction list 124 that is a list of blog article information collected from the blog site 20, and a site article introduction list 124. A keyword cloud 121 of keywords (hereinafter referred to as blog site keywords) extracted from the blog article information included in the URL, and a site related list 125 introducing blog article information in which all the blog site keywords included in the keyword cloud 121 are described. Is displayed.

  The blog site introduction information 122 for introducing the blog site 20 includes the contents of the Channel element 220 of the blog article information obtained from the blog site 20 to be introduced, and the index information 114 of the blog site 20 described above. The user can know the theme of the blog site 20 and the popularity of the blog site 20 by referring to the blog site introduction information 122.

  In FIG. 9, the site article introduction list 124 includes a maximum of 10 blog article information in the order of release date, and the title of the blog article 21 extracted from the blog article information, the release date of the blog article 21, and the blog article 21 A summary is displayed. The site article introduction list 124 may include all blog article information acquired from the blog site 20.

  In this embodiment, when extracting a blog site keyword, the target article condition is blog article information (Item element 221 and Channel element 220) included in the site article introduction list 124.

  By referring to the keyword cloud 121 on the blog site introduction page 120, the user can know the genre described mainly in the blog site 20. For example, in FIG. 9, “SFX” and “animation” are highlighted in the keyword cloud 121 of the blog site keyword. That is, it can be estimated that this blog site 20 describes articles related to “SFX” and “animation”.

  In the site related list 125 displayed on the blog site introduction page 120, as described above, blog article information in which all blog site keywords included in the keyword cloud 121 are described is extracted, and the extracted blog articles are extracted. The latest five lists of information are displayed in the same manner as the related article introduction list 115 on the related article introduction page 110.

  The user obtains information on other blog sites that publish the blog article 21 similar to the blog article 21 of the blog site introduced on the blog site introduction page 120 by referring to the site related list 125. Can do.

  Note that a link to the blog site introduction page 120 is attached to the name of the blog site 20 displayed in the list of the site related list 125, and by selecting this link, the list of the site related list 125 is displayed. A blog site introduction page 120 for introducing the blog site 20 is displayed.

  One of the titles of the related article introduction list 115 displayed on the related article introduction page 110, the site article introduction list 124 displayed on the blog site introduction page 120, or the blog article 21 displayed on the site related list 125. When is selected, an entry page 130 displaying the content of the selected blog article information is displayed. FIG. 10 is a diagram for explaining the entry page 130.

  The entry page 130 is provided in order to provide the user with detailed information on the blog article information displayed in the related article introduction list 115 or the like. As detailed information of the blog article information, the entry page 130 includes a keyword (hereinafter referred to as an entry keyword) in which only the blog article information is extracted as a target article condition in addition to the summary sentence of the blog article 21 and the blog site introduction information 122 described above. The keyword cloud 131 is displayed. By referring to the keyword cloud 131 of the entry keyword 130, the user can grasp the outline of the blog article 21 without reading the text of the blog article 21 indicated by the blog article information.

  Furthermore, the trend analysis server 1 of the present embodiment is provided with a new blog page 140 in order to provide blog article information collected most recently at the time of access. By referring to the new blog page 140, the user can obtain information on topics generated at the time of access. By selecting the new blog button 123 of FIG. 9 or the new blog button 132 of FIG. 10, a new blog page 140 is displayed.

  FIG. 11 is a diagram for explaining the new blog page 140. The new blog page 140 includes a new article list 142 that is a list of blog article information (up to 30) collected at the time of access, and keywords extracted from the blog article information included in the new article list 142 (hereinafter referred to as the new article list 142). , New keywords) is displayed.

  When extracting new keywords, the target article condition is blog article information (Item element) included in the new article list 142, and the keyword cloud 141 of new keywords includes a maximum of ten new keywords. By referring to the keyword cloud 141 of newly arrived keywords, the user can know words related to topics that have occurred near the time of access.

  The display method of the new article list 142 of the new blog page 140 is the same as the related article introduction list 115 of the related article introduction page 110.

[Trend information analysis method]
From here, the trend information analysis method according to the present invention will be described. The trend information analysis method according to the present invention includes two steps executed independently of each other, one of which is a blog article information collecting step which is a step of collecting blog article information from the Internet, and the other is This is a web page generation step of analyzing the blog article information collected in the information collection step and generating a web page for providing the analysis result to the user.

  In the network system of FIG. 1, the trend analysis server 1 executes the blog article information collection process and the Web page generation process. From here, each process contained in the trend information analysis method which concerns on this invention is demonstrated, demonstrating operation | movement of the trend analysis server 1 mentioned above.

  First, the blog article information collection process, which is a process of collecting blog article information, will be described. FIG. 12 is a flowchart showing the procedure of the blog article information collection process.

  The procedure shown in FIG. 12 is a procedure that is repeatedly executed every time a trigger is established. The trigger condition is a time condition (a certain constant when the update notification ping server 3 provides information in the PULL type). If the information is provided in the push type, the update notification ping server 3 is notified of the update of the blog update information.

  The first step S1 of this procedure is a step of obtaining the blog update information 30 published on the update notification ping server 3. The blog article information collection means 12 of the trend analysis server 1 accesses the update notification ping server 3 every time a trigger is established.

  The next step S2 is a step of acquiring the blog update information 30. The blog article information collection means 12 of the trend analysis server 1 accesses the update notification ping server 3 every time a trigger is established, and acquires the updated blog update information 30 from the update notification ping server 3. In the blog update information 30, the URL of the blog article 21 published within a certain period is described.

  The next step S3 is a step of collecting blog article information from the blog site 20. In this step, the blog article information collecting means 2 of the trend analysis server 1 accesses the blog site 20 indicated by the blog update information 30 acquired in step S2, and the meta information 22 and the blog article published on the blog site 20 are accessed. 21 link information is collected as blog article information, and the collected blog article information is stored in the information storage means 13. With this step, the procedure of the information collecting process ends.

  Next, a Web page generation process for generating a Web page for analyzing the blog article information collected in the information collection process and providing the analysis result to the user will be described. FIG. 13 is a diagram modeling the Web page generation process.

  As shown in FIG. 13, the Web page generation process includes a trend keyword process S10, a related article introduction process S11, a blog site introduction process S12, an entry process S13, and a new blog process S14. In each process, a process result is displayed. Web pages are generated, and the Web pages respectively correspond to the Web pages shown in FIG.

  That is, the Web page generated in the trend keyword process S10 is the trend keyword page 100 that displays the trend keyword, and the Web page generated in the related article introduction process S11 is the related article introduction page 110 of the selected trend keyword. is there.

  The web page generated in the blog site introduction process S12 is the blog site introduction page 120 that introduces the blog site, and the Web page generated in the entry process S13 is the entry page 130 that introduces one blog article 21. The Web page generated in the new blog process S14 is the new blog page 140 that provides information on the blog article information collected most recently at the time of access.

  The first process executed in the Web page generation process is the trend keyword process S10, and the next process executed is the related article introduction process S11. The related article introduction process S11 and subsequent steps are determined by the user's selection, and the blog site introduction process S12, the entry process S13, and the new blog process S14 are executed.

  The condition for shifting to each process is when a text in which a link to another Web page is embedded is selected. For example, the text describing the name of the blog site 20 is embedded with a link to the blog site introduction page 120. When this text is selected, the blog site introduction process S12 is executed to generate the blog site introduction page 120. The

  As described above, a keyword cloud of keywords extracted under different target article conditions is displayed on the Web page generated by each process included in the Web page generation process.

  FIG. 14 is a diagram for explaining a schematic procedure of processing for extracting a keyword. Keywords included in the Web page generated in each process of the Web page generation process are extracted by the procedure shown in FIG.

  The first step S20 of this procedure is a step in which target article conditions that define blog article information used when extracting keywords are specified. The target article condition specified in this step varies depending on the process, and by extracting a keyword from the blog article information adapted to different target article conditions, different information can be transmitted to the user in each process.

  FIG. 15 is a diagram showing an example of target article conditions for keywords. For example, in the related article introduction process S11 of FIG. 15, since the latest five blog article information including the selected trend keyword becomes the target article condition, the selection is made by referring to the keyword cloud of the extracted keyword. The reason why the trend keyword that has been selected is attracting attention can be obtained.

  In the blog site introduction process S12, since the latest 30 blog article information of the selected blog site is the target article condition, the selected blog site is referred to by referring to the keyword cloud of the extracted keywords. You can know the areas of interest.

  The next step S21 is a step of extracting keywords to be included in the keyword cloud based on the target article condition specified in step S20. As described above, since the target article condition is different in each process, the keyword extracted in this step is different for each process.

  The next step S22 is a step of generating a Web page that displays the keyword extracted in step S21 and the list of blog articles 21. The keywords extracted in step S21 are displayed in a keyword cloud in which the character size of keywords with high appearance frequency is increased. With this step, the outline procedure of the process of extracting keywords is completed.

  In addition, by including the index information 114, which is the result of evaluating the blog site, in the web page generated in the related article introduction process S11, the blog site introduction process S12, and the new arrival blog process S14, the popularity of the blog site, etc. Information can be given to the user.

  Further, in the related article introduction process S11, the transition of the ranking of the selected trend keyword is shown, and the transition graph 112 as shown in FIG. Can figure out.

The figure which showed an example of the network system which has arrange | positioned the trend analysis server. 1 is a block diagram of a network system. The figure explaining the meta information currently published by the blog site. The figure for demonstrating the web page which a web page production | generation means produces | generates. The figure for demonstrating a trend keyword page. The figure explaining a related article introduction page. The figure explaining the index information of a blog site. The figure explaining the transition graph of the ranking ranking of a trend keyword. The figure for demonstrating a blog site introduction page. The figure for demonstrating an entry page. The figure for demonstrating a new arrival blog page. The flowchart which showed the procedure of the blog article information collection process. The figure which modeled the web page production | generation process. The figure explaining the schematic procedure of the process which extracts a keyword. The figure which showed an example of the target article conditions of a keyword.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Trend analysis server 10 Keyword extraction means 11 Web page generation means 12 Blog article information collection means 13 Information storage means 2 Blog management server 20 Blog site 21 Blog article 22 Meta information 3 Update notification ping server 30 Blog update information 4 Personal computer 100 Trend Keyword page 110 Related article introduction page 120 Blog site introduction page 130 Entry page 140 New blog page

Claims (24)

A trend analysis server that analyzes information published on a plurality of blog sites on the Internet and provides an analysis result to a terminal device connected to the Internet, the trend analysis server being published on the blog site For each blog article, information on the blog site where the blog article is published and article information collecting means for acquiring the blog article title, content, and publication date as blog article information, and reference when extracting keywords Analyzing all the blog article information that meets the conditions for defining the blog article information (target article conditions), and extracting one or more words as keywords according to a predetermined criterion, and the keyword extracting means Generate a web page that displays at least the extracted keywords eb page generation means, wherein the Web page generation means displays a keyword obtained by extracting the blog article information of the blog article published within a first period as the target article condition as a trend keyword. In the blog article information in which the selected trend keyword is described when one trend keyword is selected from the Web page and the trend keyword displayed on the first Web page, The keyword (related keyword) extracted from the latest predetermined number of the blog article information as the target article condition and the blog article information referred to when the related keyword is extracted are included in the blog article information. listing by writing the name of the site, and the second of the Web page that displays, When the name of the blog site shown is selected, among the blog article information obtained from the selected blog site, the latest predetermined number of the blog article information as the target article condition, Generating a third Web page that displays a keyword (blog site keyword) extracted using the keyword extracting means and a list of the blog article information referred to when the blog site keyword is extracted ; Trend analysis server.   2. The trend analysis server according to claim 1, further comprising information storage means for storing the blog article information collected by the blog article information collection means, wherein the keyword extraction means is stored in the information storage means. A trend analysis server characterized by being a means for extracting keywords using blog article information.   The trend analysis server according to claim 1 or 2, wherein a keyword cloud of keywords extracted by the keyword extraction unit is displayed on the Web page generated by the Web page generation unit. Trend analysis server.   The trend analysis server according to any one of claims 1 to 3, wherein the blog article information acquired from the plurality of blog sites by the trend analysis server is meta information published on the blog site. Trend analysis server characterized by being all or part of it. The trend analysis server according to any one of claims 1 to 4, wherein when the blog article information referred to when the keyword is extracted is listed, in addition to the name of the blog site, of the blog post that is included in the blog post information title, you and, trend analysis server, characterized in that to write a summary statement of the blog post obtained from the blog post information.   The trend analysis server according to any one of claims 1 to 5, wherein the blog article information is acquired as an item for listing and displaying the blog article information referred to when the keyword is extracted. A trend analysis server characterized by including index information indicating a result of evaluating a blog site with a predetermined index.   7. The trend analysis server according to claim 6, wherein as the index for evaluating the blog site, the number of links to the blog site, the update frequency of the blog site, and the blog site via the trend analysis server. A trend analysis server characterized by using all or part of the number of accesses to 8. The trend analysis server according to claim 1, wherein the Web page generation unit displays a graph showing the frequency of appearance of the trend keyword and the transition of the ranking information in the second Web. A trend analysis server characterized by being included in a page. 9. The trend analysis server according to claim 1 , wherein the blog article displayed on the third Web page is displayed on the third Web page generated by the Web page generation unit. The trend analysis server, wherein the index information of the blog site is displayed separately from the list of blog article information without including the index information of the site in the information list. 10. The trend analysis server according to claim 1 , wherein the web page generation unit displays the blog displayed on the third web page when the third web page is generated. 10. The blog article information in which all site keywords are described is extracted, and the latest predetermined number of the extracted blog article information is listed and displayed on the third Web page. . The trend analysis server according to any one of claims 5 to 7 , wherein the web page generation unit selects a title of the blog article displayed in the list of blog article information. The blog article information from which the title of the blog article is selected is the target article condition, the keyword (entry keyword) extracted using the keyword extracting means, and the blog article information referred to when the entry keyword is extracted A trend analysis server that generates a fourth Web page that displays a list of only web pages. A trend analysis server as claimed in any one of claims 11, wherein the Web page generation means, wherein said first of said posts, published short said second period of time than Fifth Web for displaying the keyword (new keyword) extracted by using the keyword extracting means and the list of the blog article information referred to when the new keyword is extracted, using the blog article information as the target article condition A trend analysis server characterized by generating pages. A trend analysis method for analyzing information published on a plurality of blog sites on the Internet and providing an analysis result to a terminal device connected to the Internet, wherein the trend analysis method is disclosed on the blog site. For each blog article, the blog article information collecting step for collecting information of the blog site where the blog article is published, the content of the blog article, and the publication date of the blog article as blog article information, and extracting keywords Analyzing all the blog article information that meets the conditions (target article conditions) that define the blog article information that is sometimes referred to, extracts one or more words as keywords according to a predetermined criterion, and displays at least the extracted keywords A Web page generation step for generating a Web page to be generated, In the process, a first Web page that displays a keyword obtained by extracting the blog article information of the blog article published within a first period as the target article condition as a trend keyword and the first Web page are displayed. When one trend keyword is selected from the selected trend keywords, the latest predetermined number of the blog article information among the blog article information in which the selected trend keyword is described is the target article. keywords (related keywords) extracted as a condition, the said posts information referred to when the extracted relevant keywords, and listed by including the name of the blog site included in the posts information, first that displays and 2 of the Web page, when the selected name of the blog sites that are displayed, Among the blog article information obtained from the selected blog site, a keyword (blog site keyword) obtained by extracting the latest predetermined number of the blog article information as the target article condition, and the blog site keyword are extracted. A trend analysis method characterized in that a third Web page that displays a list of blog article information that is sometimes referred to is generated.   14. The trend analysis method according to claim 13, wherein the blog article information collecting step is a step of storing the collected blog article information in an information storage device, and the Web page generating step is a step of storing the blog article information. A trend analysis method, wherein a keyword is extracted using the blog article information stored in the information storage device in a collecting step.   15. The trend analysis method according to claim 13, wherein a keyword cloud of the extracted keywords is displayed on the Web page generated in the Web page generation step. .   16. The trend analysis method according to claim 13, wherein the blog article information acquired by the trend analysis server from the plurality of blog sites is meta information published on the blog site. A trend analysis method characterized by being all or part. 17. The trend analysis method according to claim 13, wherein the blog article information referred to when the keyword is extracted is displayed in a list in addition to the name of the blog site. , the blog post said log information is included in the article title, you and, trend analysis method characterized by the inclusion of a summary statement obtained from the blog post.   The trend analysis method according to any one of claims 13 to 17, wherein the blog article information is acquired as an item for listing and displaying the blog article information referred to when the keyword is extracted. A trend analysis method characterized by including index information indicating a result of evaluating a blog site with a predetermined index.   19. The trend analysis method according to claim 18, wherein the index for evaluating the blog site includes the number of links to the blog site, the update frequency of the blog site, and the blog site via the trend analysis server. A trend analysis method characterized by using all or a part of the number of times of access. 20. The trend analysis method according to claim 13, wherein the Web page generation step displays a graph indicating the trend frequency of the trend keyword and the transition of the ranking information in the second Web. A trend analysis method characterized by being included in a page. A trend analysis method according to claims 13 to claim 20, wherein the third Web page, the blog to be displayed in the third Web pages generated by the Web page generation step The trend analysis method, wherein the index information of the blog site is displayed separately from the list of blog article information without including the index information of the site in the article information list. The trend analysis method according to any one of claims 13 to 21, wherein, in the Web page generation step, a blog site displayed on the third Web page when generating the third Web page The trend analysis method, wherein the blog article information in which all of the keywords are described is extracted, and the latest predetermined number of the extracted blog article information is listed and displayed on the third Web page . The trend analysis method according to any one of claims 17 to 19 , wherein, in the web page generation step, when a title of the blog article displayed in the list of blog article information is selected, A keyword (entry keyword) extracted from the blog article information from which the title of the blog article is selected as the target article condition and only the blog article information referred to when the entry keyword is extracted are listed and displayed. 4. A trend analysis method characterized by generating four Web pages. 24. The trend analysis method according to claim 13 , wherein, in the web page generation step, the blog article published within the second period shorter than the first period. A fifth Web page is generated that displays a keyword (new keyword) extracted from blog article information as the target article condition and a list of the blog article information referred to when the new keyword is extracted. Trend analysis method.

Images