KR20120014466A - An blog prestige ranking method based on weighted indexing of terms - Google Patents

Abstract

PURPOSE: A blog authority ranking method based on a term weighting index is provided to improve the quality of a blog searching result by assigning weighted values to important terms. CONSTITUTION: A web robot module collects necessary data by analyzing the structure of blog posts. The weighted keyword ranking score of the blog post is calculated by assigning weighted values to a post title, a tag, and an article. A term weighting rank score is stored and calculated by collecting the normal word score the important word score of the blog post. A blog list is calculated based on the word weighting rank score of the important word corresponding to a keyword.

Description

An blog prestige ranking method based on weighted indexing of terms}

The present invention relates to a blog authority ranking method based on the term weighting index. In particular, the term weighting index for further improving the search result quality of the blog search system through the term weighting index using the structural characteristics of blog posts on the Internet. Blog based on authority ranking method.

In general, blogs are websites that record information about an individual's thoughts or interests in the form of a journal, which allows them to share opinions and information with others. Blogs give a glimpse of the interests and thoughts of people around the world and their reactions to a company's products or services. Popular blogs also see hundreds of thousands of new blog posts a day, so the ripple effect is as great as the media. In addition, real-time Social Network Services, such as Twitter, have made blog posts more powerful. For this reason, the blog search field can be a good contributor to the study of the flow or ripple of marketing, media, and Internet information.

The method for searching authority and ranking authority (hereinafter referred to as 'blog ranking method') has been studied a lot until recently. In particular, due to the characteristics of blogs that have different characteristics and structure than general web documents, blogs' ranking methods have been studied in various ways. Unlike regular web pages, blog-based authority rankings, such as page rank, are not effective because links between pages are not enough to be used as a key factor in ranking.

Due to these problems, a blog ranking method has been developed that applies ranking based on link connection by analyzing similarity and accessibility between blogs. In addition, it has been developed further, the blog ranking method was developed considering the number of new blog posts and the number of comments for each article in a certain time. Among the conventional methods, the EigenRumor search and the Egocentric search, which is further developed, use the amount of trackbacks connected between the blog post and the comment between the user or the post of the blog as the eigenvectors. The user's evaluation of each post on the blog was quantified and applied to the blog authority ranking.

In the EigenRumor blog ranking method, the hub is the authority score that quantifies how many blog posts are popular in the blog community for each blog site, and the score that bloggers rated popular blog posts. With a score, you get a Reputation score that quantifies how much of your blog posts in each blog site have received high-value bloggers.

FIG. 1 relates to a method for determining a blog ranking by a conventional method such as the EigenRumor blog ranking method. As shown in FIG. 1, a blog post comment and trackback analyzing a blog post user response as shown in FIG. Measure the authority of your blog posts by adding up the links and applying them to the ranking.

Blog search using the conventional blog ranking determination method has a high authority of blog posts, but has a low relevance to a search word, so that the quality of the search result is frequently caused. In keyword-based search, it is a good search system when the relevance between the search word and the search result is high. Therefore, the adequacy of the blog post and the search term corresponding to the user's desired result should also be reflected in the ranking determination method independently of the authority of the blog post, but the conventional blog ranking determination method does not consider this.

Despite the fact that the user's search terms are an insignificant part of the content of the blog post, if the blog post itself is high in authority, it is obvious that the user will not judge the result as desired. In addition, even if the frequency of each term in the blog post is considered in consideration of the importance of the index word, the quality of the search results is not guaranteed, thereby raising the need for a more effective method of reflecting the importance of the term in the blog post.

The present invention has been invented in view of the above necessity, and is a method and system for determining blog rankings by weighting terms using structural characteristics of a blog for an effective blog search system. By analyzing the structural characteristics of blogs, it is possible to determine the importance of terms occurring in blog posts. Through this, blog post rankings of each term can be determined and reflected in blog search results. In addition, it aims to provide a blog authority ranking method based on the term weighting index that can improve the quality of blog search results by considering the suitability of index terms and close terms.

The present invention for achieving the above object,

A preprocessing step of searching for blogs activated on the Internet through a web robot module to analyze the structure of each blog post, and collecting and normalizing data necessary for determining a blog ranking through the analyzed structure;

Ranking for calculating weighted important term ranking scores of blog posts by applying weights for each structure in terms of the post title, tag part, and important phrases in the text among the blog post data analyzed through the preprocessing step Determination step;

The term of the blog post is calculated by adding the weighted key term score in the ranking determination step and the general term score to which the frequency is applied to the rest of the general terms in the blog post body as the overall term score for the post to calculate the term weighted ranking score. A database storage step of storing the database in each database; And,

And a search step of calculating a list of blogs based on the term weighting ranking scores of blog important terms corresponding to search terms requested by a user query server.

The present invention improves the quality of blog search results by increasing the suitability of blog post rankings and terms by assigning weights to important terms using the structural characteristics of blogs in blog post rankings. When ranking blog posts, the index of each term reflecting the relevance of the blog post to each term, rather than the authority score of the entire blog post, is indexed in the blog search system database. The final ranking score reflected in the search result may provide a user with a result list of posts that are considered up-to-date among blogs that are highly related to the search word input by the user. The blog post itself is renowned and authoritative, but you can expect high suitability of the search results while avoiding the problem that the results that do not match the information desired by the user are ranked above the search results.

1 is a view for showing the authority score calculation in the conventional blog ranking determination method.
2 is a structure of a general blog post.
3 is a view illustrating a term weighting blog post ranking method according to the present invention.
4 is a schematic diagram of a blog search system to which the term weighted blog post ranking method according to the present invention is applied.
5 is a diagram illustrating a relational database schema of a blog search system according to the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a cross-sectional view illustrating a structure of a general blog post. A blog usually includes a plurality of blog posts. The blog post data includes title, category, date, etc., body data, tag parts, comments, trackbacks, and the like. It consists of. One post corresponds to one document in traditional information retrieval.

The blog ranking determination method according to the present invention first performs a preprocessing process for performing a blog and blog post data collection and analysis process. The preprocessing process searches the blogs activated on the Internet through the web robot module of the search system, analyzes the structure of each blog posts, and collects and normalizes the data necessary for blog ranking determination through the analyzed structure. To pass.

The ranking determination module calculates the ranking score for each term using the processed data as a parameter and passes it back to the blog search system. The blog search system stores the score in a database according to the relationship between the corresponding blog post and the term. This is used to output the search result that the user finds in the search query server of the search system.

In the blog ranking determination method according to the present invention, as shown in FIG. 3, the blog post data analyzed by the web robot module of the retrieval system is used for terms existing in the title, tag part, and important phrase in the text. Calculate the key term ranking score of blog posts by applying the weight for each structure.

Key term ranking scores are given the highest weight in the title, followed by tag and anchor text, respectively. In addition, the frequency of the general terms in the post body is calculated and summed with the key term ranking score to store the term weighting ranking score for each term of the blog post.

In other words, the weighted weighted key terms scored differentially weighted according to each structure are calculated for the key terms extracted through the structural characteristics of the post, and the frequency is calculated for the remaining general terms in the main text. The term weighting ranking score is stored in the database for each term of the blog post by adding the term score to the overall term score for the post. The database indexes not only posts and general terms, but also important terms.

Here, the frequency of each term in the post body sets an upper limit in order to keep the reflection on the ranking lower than the score of the weighted key term. In order to reflect the freshness of posts in the final ranking, the creation date information of each post is also collected and indexed.

Since important terms representing blog posts may correspond to user search terms that are highly related to blog posts, the weighted scores may be increased by giving higher weights than general terms. In addition, the ranking score is reflected in the search results of the search system.

Among blog structures, category is a criterion for classifying blog posts by blogger according to subject or material, so the terms used in categories can be important terms of blog posts. However, due to the limitations of pre-set generic terms, important term weights are set lower than those of post titles or tags. Post titles are most relevant to the topic of the post, so give them a higher weight. In addition, in case of a tag in which a user inputs a subject, material, or classification of the post body, the weight is given lower than the title. In the case of categories and tags, there are many cases that are composed of terms that have nothing to do with the post body, depending on the usage pattern of the blogger. In addition, there are subheadings or anchor texts where structurally important terms can be extracted from the text. The anchor text is the description part of the link link to cite an external web document in the text. This is also an important term, and the weight is given, and weights are also differentially applied to terms extracted from bold text, underline, and accent color.

As shown in FIG. 4, the blog search system 100 to which the method of the present invention is applied weights a web robot 110 that collects necessary data from a blog and a term that ranks the collected data by terms and blog posts. Ranking unit 120, the database 130 for indexing and storing the ranking scores and blog information, the user query server 140 that finds the search results according to the user's request in the database 130 and outputs the ranking in high order It consists of.

The database 130 stores basic information of a blog post (title, date, blogger name), a list of terms corresponding to all terms existing in the blog post, and an authority score and a term score for each blog post. Are stored respectively. In addition, all important terms of a post belonging to the same blog are indexed separately by blog.

5 is an illustration of the overall schema when indexing such a database into a relational database system. The information stored in the database is primarily calculated based on the ranking scores of the important blog terms linked to the terms corresponding to the search terms requested by the user query server. The scores for the final result are also calculated by summing the numbers. The user query server retrieves blog post information corresponding to the list of blog posts in the order of highest final ranking score from the database and outputs the search result as a search result.

In addition, in another embodiment of the present invention, the database 130 provides the user query server 140 with a freshness factor index obtained by converting blog generation date recording information into a normalized number, and the user query server 140 provides the user with the index. When the search request is inputted, the final ranking score is calculated and provided to the user query server using the key term score and the latestness factor corresponding to the search word. The user query server 140 firstly calculates a list of blogs having a high term ranking score and a high relevance factor among blogs in the database 130 in which a term corresponding to a search term requested by the user is related to the key term index. Actually calculates the ranking list of posts related to the search word entered by the user in the second and finally provides the result divided into the first and second stages.

Therefore, in the blog authority ranking method based on the term weighting index according to the present invention, and a system for performing the same, a result list of posts considering the most up-to-date among blogs that are related to a search word entered by a user is taken from a database. You can calculate the final blog ranking score that you can provide, so that the blog post itself is well known and authoritative, but that results that do not match the information you want are ranked high in search results, while avoiding the problem of ranking above the search results. You can expect.

Claims (3)

Iii) a preprocessing step for searching the blogs activated on the Internet through the web robot module to analyze the structure of the posts of each blog, and collecting and normalizing data necessary for determining the blog ranking through the analyzed structure;
Ii) The weighted key term ranking score of the blog post is calculated by applying weights for each structure among terms existing in the title, tag portion, and important phrases of the post among the blog post data analyzed through the preprocessing step. Ranking determination step for;
Iii) the weighted key term score in the ranking determination step and the general term score applied with the frequency for the remaining general terms of the blog post body are added to the overall term score for the post to calculate the term weighted ranking score blog A database storage step of storing the database in terms of post terms; And,
Iii) a search step of calculating a list of blogs based on the term weighting ranking scores of the important blog terms corresponding to the search terms requested by the user query server based on the information stored in the database. The weighted key term score in the database storing step is the highest in the title, and is sequentially given in the order of tag and anchor text, and the weights are differentially applied to terms extracted from bold, underline, and accent color. The terminology weighting index based blog authority ranking method, characterized in that for setting the upper limit lower than the weighted key term score is applied to the general term score applied frequency. The method of claim 1, wherein the retrieving step calculates a final ranking score using a recency factor obtained by converting the term weighting ranking score and blog creation date record information into a normalized value, thereby obtaining a high term weighting ranking score and relativity factor. A blog authority ranking method based on a term weighted index, wherein the blog list is first generated, and a list of posts related to a search word actually input by a user is calculated and provided second.

Images