KR100987284B1

KR100987284B1 - Apparatus and method of creating white-list of search-engine

Info

Publication number: KR100987284B1
Application number: KR1020080090174A
Authority: KR
Inventors: 권오성
Original assignee: 주식회사 엔씨소프트
Priority date: 2008-09-12
Filing date: 2008-09-12
Publication date: 2010-10-12
Also published as: KR20100031198A

Abstract

The present invention relates to an apparatus and method for generating a white list of a search engine, and more particularly, comprising: a domain information collecting unit for recommending domains for each subject on a plurality of subjects from a plurality of clients through the Internet; And generating a tree structure using a plurality of topics and one or more domain information belonging to each topic, wherein the nodes of the tree structure may include one domain information and recommendation information for the corresponding domain, and include the domain information. Is a sub-node of the subject to which it belongs, and between the sub-nodes belonging to the same topic, the tree structure generation unit to determine the layer according to the recommendation information; characterized in that, according to the present invention the search engine is a high search It is possible to maintain the quality of the results, and netizens are provided with incentives to author high quality content.

Whitelist, tree structure, domain,

Description

Apparatus and method of whitelist generation of search engines {APPARATUS AND METHOD OF CREATING WHITE-LIST OF SEARCH-ENGINE}

The present invention relates to a method for constructing and updating information based on criteria and basis used by a search engine to generate search results of quality content.

Since the advent of the so-called Web 2.0 era, much of the content generated on the Internet has been authored directly by netizens, including a mix of end-user experience and expert knowledge. If such high-quality content can be selected and returned as a search result, there can be no doubt that it will be a quantum leap in comparison with the prior art, which simply returns random web pages and contents related to the search word.

In particular, the number of web pages present on the Internet is already exploding to the extent that it is difficult for search engines to index, and indexing without obscuring any web pages is not worth the information or even low-value web pages. By including them as targets, the results may be contrary to the improvement of the quality of the search results, and the concept of returning to the search results in consideration of sites having "quality content" in the first place can lead to a revolutionary paradigm shift. will be.

However, without adopting the rhetoric of Web 2.0, the adoption of this methodology has provided incentives for netizens working on different sites to author quality content, and how high quality a site can be on a particular topic. It is necessary to prepare a method or a measure to determine whether the content is held.

The present invention has been made to solve the above problems of the prior art, and to provide an apparatus and method for generating a white list of a search engine for encouraging the improvement of the search result quality of the search engine and the active authoring of content by netizens. do.

In order to achieve the above object, the apparatus for generating a white list of a search engine of the present invention comprises: a domain information collecting unit for recommending domains for each subject on a plurality of subjects from a plurality of clients through the Internet; And

A tree structure is created by using a plurality of topics and one or more domain information belonging to each topic, and the nodes of the tree structure may include one domain information and recommendation information about the domain. And a tree structure generation unit that becomes a subordinate node of the subject to which it belongs, and determines the level according to the recommendation information among the subordinate nodes belonging to the same subject.

Meanwhile, in order to achieve the above object, the method of generating a whitelist of a search engine of the present invention collects information on correlations between respective topics and domains by receiving domain recommendations from a plurality of clients for each subject on a plurality of topics. Making;

The tree structure is created by using the collected information, and the nodes of the tree structure may include one domain information and recommendation information for the corresponding domain. Generating a tree structure between the lower nodes belonging to the same subject to determine the level according to the recommendation information; And

And receiving a query from a client, obtaining a search result by referring to a tree structure, and returning the search result.

According to the present invention as described above there is an effect that the domains included in the white list is motivated to have a higher quality content.

In particular, each netizen used in the domain can be actively engaged in the domain by giving the search engine the opportunity to easily expose their authoring activities to the unspecified number of users. In addition to acting as an incentive to do so, it can directly induce reconsideration as a speaker in the entire Internet space, leading to active netizen participation appropriate for the Web 2.0 era.

In addition, due to this synergy, domains have higher quality contents, and by using a white list including these domains, a search engine can maintain high quality search results.

Hereinafter, a configuration of an apparatus for generating a white list of a search engine according to the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a reference diagram exemplarily illustrating a structure of a white list, and FIG. 2 is a functional block diagram showing a white list generating apparatus of a search engine according to the present invention.

The dictionary meaning of a whitelist is "a list of preferred items in the whole" and is used as the opposite concept of a blacklist, which usually means a list of unfavorable or rejected items.

Meanwhile, in the present invention, the white list is data of a tree structure in which each node has a specific domain address as a value, and the domain of the node located at the top of the tree structure is evaluated to include valuable information related to a specific search word or a specific topic. Say that.

FIG. 1 shows an example of a white list according to the present invention. Since the root node has no value, the root node may be represented by four list structures when the root node is removed. It will be collectively referred to as a tree structure.

On the other hand, in Figure 1, the root node is connected to four sub-nodes, each link from the left has a value of "news", "shopping", "blog", "hobby", respectively, by the "news" link Linked subnodes have a value of "next", subnodes linked by a "shopping" link have a value of "auction", subnodes linked by a link called "blog" have a value of "tistory", link "hobby" It can be seen that the child node connected by has no value.

In this case, the values "news", "shopping", "blog", and "hobby" that a link has means a subject to which a child node belongs, and "next", "auction", and "tistory" which each node has, respectively. Refers to the domain name. In the example of FIG. 1, a letter is shown for clarity, but the domain means a domain name registered in the International Network Information Center (InterNIC) or the Korea Internet Information Center (KRNIC). In the example, "www.daum.net" , "www.auction.co.kr" and "www.tistory.com".

On the other hand, "hobby" is divided into four sub-topics, "game," "picture", "car", and "programming", respectively, that is, subjects within a white list may have a hierarchical structure. The subnode linked by the third link in "Car" has a value of "baby dream (www.bobaedream.com)", while the "baby dream" node is again a child of "www.encar.com". As a node, the "enka" node is connected to a subnode called "Naver car" (car.naver.com).

The higher the fidelity is in the subject to which the node belongs to the domain belonging to the node above the white list. In the example of FIG. 1, in the subject “hobbies-cars”, the “baby dream” node is the highest node, the “enka” node is the next node, and the leaf “Naver car” is the lowest node. This means that in the "Hobby-Car" theme, the "baby dream" domain (www.bobaedream.com) is rated as having the best quality information about "Hobby-Car".

In other words, it is more likely that "baby dream" web pages will contain better quality information than the web pages of the leaf "car.naver.com".

Meanwhile, according to FIG. 2, the apparatus 100 for generating a white list of a search engine according to the present invention includes a domain information collecting unit 110, a tree structure generating unit 120, and a search processing unit 130.

First, as shown in FIG. 3, the domain information collecting unit receives recommendation from the netizens about which domains have the best content about the respective subjects. Netizens participate in the domain information collection unit 110 through the Internet using a client such as a personal computer, by recommending domains for each subject, or voting for domains recommended by other netizens. Although not represented in FIG. 3, the topic itself may be recommended, or keywords related to the topic may be recommended.

Meanwhile, the tree structure generation unit 120 generates a tree structure (white list) as shown in FIG. 1 by using the information collected by the domain information collecting unit 110, that is, information about domains recommended for each topic. .

In this case, each node of the tree structure may include one domain information and recommendation information (information indicating how many recommendations have been received) for the corresponding domain. Meanwhile, as in the root node of FIG. 1, there may be a node having no subject information or a subject such as "car" or "hobby" that is not domain information.

On the other hand, a node including domain information and recommendation information about the domain becomes a sub node of a subject to which each belongs. In the example of FIG. 1, the topics are stored in links instead of being stored in nodes, but whether the topics are stored in links or in separate nodes, these topics each include nodes with subdomains whose domain belongs to them. .

In addition, each node in the generated white list is located in the white list according to recommendation information about domain information that it includes, that is, how many netizens recommend the corresponding domain. Preferably, as shown in the example of FIG. 1, a domain that has been highly recommended is determined as an upper node, and a domain that has not been recommended as a lower node is determined.

Meanwhile, the search processor 130 performs a function of a conventional search engine that receives a query from a client and returns a search result. The exposure order of the search result is generated by using the white list generated by the tree structure generator 120. Determine and return

That is, the search processing unit 130 of the search results so that the web page or the content belonging to the domain included in the white list of the search results, if the web page or content included in the domain belonging to the parent node more easily By reorganizing the search results, such as at the top, it makes it easier for netizens to find domains that are rated as having good information about a particular subject.

Meanwhile, as illustrated in FIG. 3, the tree structure generation unit 120 generates a first white list by receiving recommendations from netizens, and then updates the tree structure through feedback of netizens on the search result of the search processing unit 130. do. Recommendations for a particular domain, which are the information contained by the nodes in the tree structure, is an indicator of how good the domain has information about the subject to which the node belongs. This is information that can be changed dynamically. We need to re-evaluate to raise the position of the node up to make it a higher node, or rather to make it a lower node.

The updating of the tree structure may be based on the frequency of netizens clicking on a link and moving to the web page. Alternatively, the tree structure may be separately evaluated by the netizens and reflected.

On the other hand, using the white list to expose the domains belonging to the higher priority nodes as search results may be an incentive for each domain to secure quality information in addition to improving the quality of the search results.

Furthermore, the more active the netizens working in the domain, the more frequently the articles, articles, photos, videos, etc., written by the user are exposed to the unspecified number through the search engine, thus acting as an incentive.

As such, in order to encourage netizens who are active in each domain to create more active content, the search processor 130 may identify the author for each search result as shown in FIG. 4 when the search result is exposed, and You can also display evaluation information about the author.

In general, web sites where content is created by members are generally operated as a member system, and each member uses an ID to identify himself / herself. The identification information is preferably an ID used in each domain. As the evaluation information, the activity amount for the ID can be used most easily, and other netizens' evaluation of separately exposed search results, or the frequency of the author's contents exposed as search results or the search results are accessed. In some cases, it may be used. In the example of FIG. 4, it can be confirmed that the ID of the author and the evaluation information of the corresponding creator are displayed for each search result through a star shape.

Hereinafter, a method of generating a white list of a search engine according to the present invention will be described with reference to FIG. 5. 5 is a flowchart illustrating a time series of a search method using a white list according to the present invention.

First, as shown in FIG. 3, domain topics are received from netizens for each subject for a plurality of subjects, and information on correlation between the subjects and domains is collected (S110).

Subsequently, a white list of a tree structure is generated using the collected information. In this case, the node of the white list may include one domain information and recommendation information for the corresponding domain. Meanwhile, the node including the domain information belongs to the subject to which it belongs (ie, becomes a subnode), and the layer is determined between nodes belonging to the same subject according to the recommendation information (S120).

Subsequently, when a query is received, a search result is obtained by referring to a white list and then returned (S130). As shown in FIG. 4, identification information for identifying an author for each search result and evaluation of the corresponding author are provided. Information can also be displayed together. Furthermore, the evaluation information of the authors may be updated by analyzing netizens' evaluation of the search results, the frequency of the contents created by the corresponding authors in the search results, or the frequency accessed by the search results.

Meanwhile, the whitelist is dynamically updated through the feedback on the search result (S140). As a result, the quality of the search results can be maintained at a high level.

Although the present invention has been described in detail with reference to several embodiments, the present invention is not limited to these embodiments and is not construed, but may be freely modified and implemented within the scope of the technical idea described in the claims.

1 is a reference diagram exemplarily illustrating a structure of a white list.

2 is a functional block diagram showing an apparatus for generating a white list of a search engine according to the present invention;

3 is a screen example illustrating a web page where a domain information collecting unit receives a domain recommendation for each subject by netizens.

4 is an exemplary screen illustrating an example of displaying identification information for identifying an author and convenience information on the corresponding author in a search result.

5 is a flowchart illustrating a time series of a search method using a white list according to the present invention.

110: domain information collecting unit 120: tree structure generation unit

130: search processing unit

Claims

A domain information collecting unit for recommending a domain for each subject on a plurality of subjects from a plurality of clients through the Internet; And

A tree structure is created by using a plurality of topics and one or more domain information belonging to each topic, and the nodes of the tree structure may include one domain information and recommendation information about the domain. And a tree structure generation unit which becomes a sub node of a subject to which the subject belongs, and determines the level according to the recommendation information among the sub nodes belonging to the same subject.

The method of claim 1,

And a search processing unit which receives a query from a client and returns a search result, and determines an exposure order of the search results using the tree structure.

The method of claim 2,

Apparatus for generating a white list of a search engine, characterized by updating a tree structure through feedback on a search result.

The method of claim 3, wherein

Feedback to the search results is the client's access frequency to each link included in the search results or the client's evaluation of the search results, characterized in that the white list generating apparatus of the search engine.

The method of claim 2,

And, when the search processing unit returns a search result, identification information for identifying the author for each search result is displayed together.

The method of claim 5,

In addition to the identification information of the author, the whitelist generation apparatus of the search engine, characterized in that to display along with the evaluation information for the author.

The method of claim 6,

And the evaluation information is updated according to the evaluation of the client about the search result, the frequency of exposure to the search result, the frequency of access by exposure to the search result, and the amount of activity in the domain.

The method of claim 1,

And a plurality of subjects have a hierarchical structure.

Collecting information on correlations between each subject and a domain by receiving domain recommendations from a plurality of clients for each subject on a plurality of subjects;

The tree structure is created by using the collected information, and the nodes of the tree structure may include one domain information and recommendation information about the domain. The node including the domain information becomes a child node of the subject to which it belongs. Generating a tree structure such that a layer is determined according to recommendation information among lower nodes belonging to the same topic; And

The method of claim 9,

Updating the tree structure through the feedback on the search results; Method of generating a white list of the search engine, characterized in that it further comprises.

The method of claim 9,

When the search results are returned, identification information for identifying the author and evaluation information for the corresponding author are displayed for each search result.

The method of claim 11,

Updating the evaluation information for the author by analyzing the client's evaluation of the search results, the frequency of exposure to the search results, or the frequency accessed by being exposed to the search results; whitelist generation of the search engine further comprising Way.