KR20050114576A - Search system and method using a plurality of searching criterion - Google Patents

Abstract

The present invention relates to a search system using a plurality of search criteria and a search method thereof.

In the present invention, when a search word is provided from a user terminal through a network, search targets for searching are set based on the search word. In addition, a search is performed by applying different search criteria to each search target, and a search result is provided for each search target.

According to the present invention, more reliable information can be provided.

Description

Search system and method using a plurality of searching criterion}

The present invention relates to a search system and a method thereof, and more particularly, to a system for performing a search using a plurality of search criteria for information provided through a network and a search method thereof.

Recently, with the rapid increase of the Internet, electronic commerce for the whole world is being made through the same medium of the Internet, and the communication function between users is gradually developed, and various forms of communication are being made through the Internet. .

In addition to e-mail or file transfer, which is the most basic form of communication using the Internet, an Internet phone is used instead of a landline phone, and a short message can be transmitted to a mobile phone using the Internet. In addition, there is also a multi-party chat using the Internet, and in the case of a chat, not only using a text, but also a voice chat or a video chat through the Internet. In addition, a community in which a large number of users gather on the same topic and exchange materials and opinions is being activated.

The community consists of various forms such as cafes, blogs, homepages, clubs and mini homepages, and various contents are posted in each community. In general, each user who is registered as a member in a cafe or club posts content on a bulletin board on a bulletin board and communicates with each other.In blogs and mini homepages, the user who opened it posts his or her opinions or materials. do.

Conventionally, in order to view or receive content posted to such a community, there is a hassle of accessing the community and searching for the content. However, recently, due to the development of a search technology, a user can search for content on a web browser. Only by inputting a corresponding search word, a content search for the search word is performed for each item such as a web document, news, image, dictionary, etc. on the web, and information corresponding to the search word is searched for in a community such as a cafe or a blog. A search for a community having a branch, and a process of searching for a content corresponding to the search word among contents (eg, posts) posted in the communities, are performed. Accordingly, the user can search for predetermined content posted in the community without having to access the community individually, and can know the community having information about the search word.

However, in the related art, since a search is mainly performed only according to whether a search word is included, there are many contents that do not include information for reference among contents provided as a search result for a community. That is, conventionally, when searching for a community, a community including a search term entered is searched, and a post including the search term is searched among posts of the community, and a plurality of content classification menus ( For example, a search is performed only according to whether a search word is included, such as searching a classification menu including the search word among bulletin boards and the like.

As described above, although the search conditions of different criteria are required according to the characteristics of the search target such as community search, post search, classification menu search, etc., since the search is performed according to the same conditions, each search has reliable information. It is hard to find things.

Therefore, the technical problem to be achieved by the present invention is to provide a reliable search results for each search target by performing a search based on different criteria for each search target, in order to solve the conventional disadvantages.

In addition, the technical problem of the present invention is to be able to change the order in which the search results for each search target is arranged, so that it is possible to easily check the reliable information about the search word input by the user.

According to an aspect of the present invention, a method of searching for information on a search word provided from a user terminal is connected to a plurality of user terminals through a network. If a search term is provided from a user terminal, setting search targets to be searched based on the search term; b) searching for information including the search word for each search target; c) calculating reliability based on different search criteria with respect to the information searched for each search target; d) determining an order in which search targets are arranged according to the search word; And e) arranging search objects according to the arrangement order, and arranging searched information for each search object according to the measured reliability and providing the search results to a user.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention.

In an embodiment of the present invention, a search is performed according to different criteria for each search target.

Here, the search target is information to be searched, and is divided into a unit, a post, and a classification menu in the present embodiment according to the form of the search, but is not limited thereto. The unit represents a community such as a cafe, a blog, a talk square, and the like, which provides content on a predetermined topic, and a post is content posted to the community. In addition, the classification menu may indicate different subject categories in the community, and may be, for example, names of bulletin boards that classify contents provided in the community. In addition, it may be a diary or photo album.

In order to provide reliable information in consideration of the characteristics of the search object, the present invention applies different search criteria to each search object. In detail, when searching for a unit, the search is performed according to a first search criterion that considers at least one evaluation variable among community user confidence index, community activity index, and community size. The search is performed according to a second search criterion that considers at least one evaluation variable among content usage, additional information attached, and size of the content. In addition, when searching a classification menu, a search is performed according to a third search criterion that considers at least one of whether a search word is included in the classification menu title and the number of posts.

In particular, the present invention measures the reliability of the search object according to each search criterion, and provides a search result according to the measured reliability, thereby providing more reliable information.

First, evaluation variables considered in the first search criterion will be described.

The community user trust index considered in the first search criterion is the sum of the trust indices of all users registered in the community, and each trust index indicates how reliable information the user provides. This user's confidence index can be measured according to the user's network activity. For example, network activity may represent the use of the Internet to answer a question posted by a user, such as in a knowledge sharing service, in which case the user's confidence index answers how many questions the user provided, It can be calculated by how much has been adopted (answer adoption rate). In addition, network activity is not limited to this knowledge sharing service, and includes all activities that provide information on the network. The user's confidence index may also include a sum of reward means (eg, mileage, etc.) provided from the system by the result of the activity.

In addition, the community user confidence index may be a value measured according to the degree to which other users used the content. For example, the confidence index of the first user is measured according to a value (eg, the number of scraps) indicating how much the second users scraped the content posted by the first user, and the sum of the confidence indexes of the first users is measured. The community user trust index may also be calculated.

In addition, the community activity index considered in the first search criterion indicates how much the community has exchanged content, and may be calculated according to the sum of activity indices of the community members, the number of visitors, and the number of relationship formation. The activity index of a member may be measured according to the community activity of the user. For example, community activity refers to a user posting content to a community, in which case the user's activity index depends on how much new content has been posted within a set period of time (for example, current based). Can be. In addition, the community activity index may vary depending on the number of users visiting the community. This is based on the fact that as users of the community provide more useful and reliable contents, other users visiting the community to which the user belongs may increase.

In addition, in the case of a personal community, the community activity index may vary according to the number of other communities having a predetermined relationship with the community. Generally, in a personal community such as a blog, like a "neighbor blog", another blog (second blog) may form a predetermined relationship to a given blog (first blog) as a neighbor. When a neighbor blog relationship is formed, a link for accessing the first blog is displayed on the second blog, so that users connected to the second blog can easily access the first blog without a separate search process. As users tend to engage with more useful and reliable blogs and neighbors, the community activity index may vary depending on the number of other communities (relationships) that have a given relationship.

As such, the community activity index may vary depending on at least one of the sum of the activity index of members, the number of users (visitors), and the number of relationship formations of the community, but is not limited thereto.

In addition, the size of the community considered in the first search criteria may vary depending on the number of users registered in the community, the number of contents posted in the community, and the like.

In the present invention, when performing a search in consideration of the first search criteria, the search is performed considering at least one of the community user confidence index, the community activity index, and the community size as described above, and the reliability of the corresponding community is measured. However, it is preferable not to measure community reliability based only on community size, and it is preferable to measure reliability by including at least one community user confidence index and community activity index.

Next, evaluation parameters considered in the second search criteria will be described.

The characteristics of the user who posted the content considered in the second search criterion indicate age, occupation, region of residence, trust index, and the like. The purpose is to share the credibility of the content based on the statistical results of posting. Therefore, the content is classified according to the user's age, occupation or region of residence, and the reliability of the information included in each classified content is evaluated, and information of users of any age or occupation is more reliable according to the evaluation result. The process of deciding whether or not to

A user's confidence index is a measure of how reliable the system posts to each user. The user confidence index may be a value measured according to the degree to which other users use the content. For example, a value (eg, number of scraps) indicating how many second users scraped content posted by a first user, or how many second users form a predetermined relationship with a community of the first user. The reliability of the first user may be measured based on the value (eg, neighbor number, etc.). In addition, the user confidence index may be a value measured according to the user's network activity. For example, network activity may represent the use of the Internet to answer a question posted by a user, such as in a knowledge sharing service, in which case the user's credibility is how many answers the question and how much It may be calculated by the fact that it is widely adopted (answer adoption rate).

Meanwhile, the content utilization considered in the second search criterion indicates how many other users have inquired and used the corresponding content. Variables that can measure how many times a user has viewed and used the content include the number of times the user has recommended the content to other users (recommendation), the number of times other users have posted replies to the content (the number of replies), Number of comments posted (number of comments), number of times the content was sent by email (number of emails sent), number of times the content was recommended to other users (recommended), number of times the content was printed (e.g., printed, etc.) (Number of outputs). Here, the comment indicates that the reply to a specific post can be viewed while viewing the post, and may be called by various names (eg, footer, comment, single line answer, etc.) for each service provider. In addition, the number of contents scraped by a user (hereinafter, referred to as "number of scraps") may be included. Scrap on a network represents providing (copying, etc.) some or all of the content posted on the network to a predetermined location. In addition, the number of executions of all actions that may be generated by a user clicking on content may be included.

In addition, whether the additional information of the content considered in the second search criterion is attached indicates whether or not additional information such as an image, a video, and audio is attached to the content in addition to the text information.

In addition, the size of the content considered in the second search criterion indicates how much information the content has, such as the number of bytes of the content. This is based on the fact that the larger the content, the more information it contains.

In the present invention, when the search is performed in consideration of the second search criteria, the search is performed by considering at least one of evaluation variables such as the characteristics of the user, the degree of use of the content, whether additional information is attached, and the size of the content. And measure the reliability of the post.

Meanwhile, the evaluation variables considered in the third search criterion may include whether a search word is included in the corresponding classification menu title, the number of posts, and the like. In addition, content utilization among the evaluation variables considered in the second search criterion described above may be used. have.

The evaluation variables for each search criterion described above are not limited to those described above.

In addition, although the present invention applies different search criteria for each search target by unit, post, and classification menu, different criteria may be applied to one search target. For example, when searching for a unit, a first search criterion or a second search criterion may be applied according to a community type. That is, when searching a public community such as a cafe, a search may be performed according to a first search criterion, and when searching a personal community such as a blog, a search may be performed according to a second search criterion.

Communities searched by these evaluation variables may include all types of communities that can occur on the network. For example, a public community (e.g., cafes, clubs, etc.) where multiple users join as members and exchange materials or opinions on the same topic, and individuals manage and post their own materials or opinions. Personal communities (blogs, personal homepages, mini homepages, profiles, etc.) may be included.

In addition to the posts posted to the public or private community, the content according to the present invention includes contents belonging to all categories (eg, web documents, news, dictionaries, etc.) provided on the network.

Next, a detailed structure and operation of a search system through a network according to the present invention having such a feature will be described.

1 illustrates a structure of a search system (hereinafter, referred to as a search system) through a network according to an exemplary embodiment of the present invention, and FIGs. 2 and 3 illustrate specific structures of each server.

Search system 100 according to an embodiment of the present invention is to provide a search service to users through the Internet, wireless communication network, etc., as shown in Figure 1, the network 200, telephone network, Internet, wireless communication network Etc.) and a plurality of user terminals 310, 320, 330, hereinafter referred to collectively as "300" for convenience.

The user terminal 300 is a communication device that can be connected to the search system 100 through the network 200. For example, various communication devices such as a wired telephone, a mobile communication terminal, a computer, and a TV with Internet access may be used. have. Therefore, the user can access the search system 100 conveniently and quickly by using PC communication, the Internet, the wireless Internet, a telephone network, and the like.

The system 100 connected to the user terminal 300 and providing a search service to a plurality of users includes an interface server 10, a community server 20, and a search server 30.

The interface server 10 allows a plurality of terminals 300 to be connected through the network 200, in particular, the Internet or the wireless Internet. In addition, a variety of information provided through each server (20, 30) is converted to meet the communication standards provided to a plurality of terminals 300, or receives the information transmitted from the terminal 300 through the network 200 It functions as a kind of web (WEB) server or WAP (WAP) server to perform the function provided to each server (20, 30).

As shown in FIG. 2, the community server 20 is a server for providing a community-related service. As shown in FIG. 2, the community server 20 provides a first community server 21 for providing and managing a public community, and a second community server for providing and managing a personal community. 22).

Each community server 21 and 22 includes databases for storing information about each community, and a processing unit for forming and managing communities among users based on the information stored therein.

Specifically, the first community server 21 stores the first processing unit 211 that performs public community formation and content management, and the general information indicating what characteristics the corresponding community has for each generated public community is stored. It includes a registration database 212, a member database 213 that stores information about users registered as members for each public community, and a content database 214 that stores content for each public community.

The registration database 212 stores rough information about the public community registered by category. For example, the identification code of the registered public community is correspondingly stored for each category code assigned to each category such as broadcasting, entertainment, economy, and politics. For each public community identification code, community information such as a public community name, a description of what characteristics, public community moderator information (such as an ID or nickname), the number of members, and a registration date are stored, and for a predetermined period (for example, Recently, information such as the number of visitors to the community is stored.

The member database 213 stores information (also known as member information) about users constituting each registered public community. For example, IDs or nicknames of users who are registered as members corresponding to the public community identification code, or member information such as password, age, gender, occupation, and e-mail address are stored, and trust index is stored for each user. It is.

The content database 214 stores information about content provided for each public community. For example, information such as a post number of a content, an ID of a content publisher, a content publication date, a content title, and content content corresponding to a public community identification code is stored. In addition, for each content, the number of replies, the number of comments, the number of times sent by mail, the number of recommendations, and the like are stored. In addition, the number of contents (hereinafter, referred to as the number of recent contents) posted in a predetermined period (for example, recently) is stored for each public community. Such content information may be stored for each category menu created in each public community. For example, as described above for each category menu of a bulletin board, diary, etc. of a predetermined public community, content information is stored.

The first processor 211 performs public community formation and management based on the databases 212 to 214 in which a plurality of pieces of information are stored. Specifically, a public community is formed and registered, user authentication is performed for each public community, and content posting and management is performed for each public community. In particular, the first processing unit 211 is a visitor monitoring module 2111 that counts the number of visitors (except users who are registered as members of the community) to access a predetermined public community, and registers as members for each predetermined public community. And a content posting monitoring module 2112 that checks the content posting of the registered user and counts the number of recent contents for each content posting.

Meanwhile, the second community server 22 may be referred to as a "blog server" as a server that provides and manages a personal community for each user. As shown in FIG. 2, the second community server 22 may include a second processing unit 221 which performs personal community formation and content management, a registration database 222 that stores information on registered personal communities, and the like. It includes a content database 223 that stores the content posted by users for each individual community.

The registration database 222 stores the general information about the personal community registered by category. For example, an identification code of a personal community registered for each category is correspondingly stored, and for each individual community identification code, a community such as a personal community name, an ID or nickname of a user who has established a personal community, a trust index of the user, a registration date, etc. The information is stored. In addition, for each individual community, information such as the number of visitors who visited the corresponding community in a predetermined period (for example, recently) and the number of relationship formation of the community is stored.

The content database 223 stores information about content provided for each individual community. For example, information such as a post number, a content posting date, a content title, and a content content of the content is stored corresponding to the personal community identification code. In addition, the number of replies, the number of comments, the number of e-mails, and the number of recommendations for the corresponding content are stored for each content. In addition, the number of recent contents posted in a predetermined period (eg, recent) for each individual community is stored, and the activity index of the corresponding user, which can be calculated based on this, is stored. Such content information may be stored for each category menu generated in each individual community. For example, as described above for each category menu such as a diary, a guestbook, and the like of a predetermined personal community, content information is stored.

The second processing unit 221 forms and registers a personal community, and performs content posting and management for each individual community. In addition, the second processing unit 221 may include a visitor monitoring module 2211 that counts the number of visitors (except users who established the community) accessing a predetermined personal community, and publishes the content of the user for each predetermined personal community. And a content posting monitoring module 2212 that checks and counts the number of recent content for each content posting. In addition, each individual community further includes a relationship formation monitoring module 2213 that counts the number of relationship formation as the visitor requests a predetermined relationship with the community. In the present embodiment, only the second processing unit 221 managing the personal community is described as including a relationship formation monitoring module. However, the present invention is not limited thereto, and the first processing unit 211 managing the public community is also the same as the relationship formation monitoring module. The module may further include a module that performs a function.

Meanwhile, in the present embodiment, the first and second community servers 21 and 22 have modules for monitoring content posting, visitor connection, and relationship formation to the corresponding community, respectively. However, such monitoring modules 2111, 2112, 2211, 2212 and 2213 may not be included in the community servers 21 and 22 and may be formed separately.

Each of the processing units 211 and 221 having such a structure increases the number of replies or comments of the corresponding content when a user other than the user who posted the content adds a reply or a comment. In addition, when another user recommends the content to another user, the number of recommendations for the content is increased.

In addition, the first community server 21 and the second community server 22 may share content by interworking with each other. For example, the user may request to publish content posted in the public community to his or her personal community through a menu such as "content addition", and in this case, the first processing unit 211 of the first community server 21. ) Copies the corresponding content to the second community server 22, and the second processing unit 221 of the second community server 22 posts the content to the user's personal community. At this time, when the content is posted on a personal community, source information of the content may be exposed.

As shown in FIG. 3, the search server 30 performing a community search according to a search word provided from the user terminal 300 measures the reliability of the search unit 31 and the searched community. The reliability measuring unit 32, a search control unit 33 for controlling the operation of the search unit and the reliability measuring unit, and a search result providing unit 34 for providing the user with the searched content according to the measured reliability.

The search unit 31 searches for a search target having information that matches the search word. That is, the search unit 31 may include a first search unit 311 and a search term for searching for a community having a topic including a word that matches a search word or having a content including a word that matches the search word. And a second search unit 312 for searching a post, and a third search unit 313 for searching a classification menu having a title including the search word or a post.

The reliability measuring unit 32 measures the reliability of each search target searched by the searching unit. Here, the reliability measurer 32 measures the reliability of the search target based on the first search criterion 321 and the second reliability measure measures the reliability of the search target based on the second search criterion. The unit 322 includes a third reliability measurer 323 that measures the reliability of the search target according to the third search criteria.

4 shows the structure of each of the reliability measuring units 321 to 323.

The first reliability measurer 321 may include a confidence index measurement module 321a for measuring a community user confidence index, an activity index measurement module 321b for measuring a community activity index, a scale measurement module 321c for measuring a community size, It includes a reliability calculation module 321d for measuring the reliability for the community based on the results of each measurement module (321a to 321c).

In addition, the second reliability measurer 322 determines whether the property measurement module 322a measures the characteristics of the user who posted the content, the utilization measurement module 322b that measures the usage of the content, and whether the additional information is attached to the content. The additional information measuring module 322c to measure, and a size measuring module 322d to measure the size of the content, the reliability calculation module for measuring the reliability of the corresponding content based on the results of each measurement module (322a to 322d) ( 322e).

The third reliability measurer 323 includes a post measurement module 323a for measuring the number of posts in each category menu, and a reliability calculation module 323b for measuring the reliability of the corresponding content based on the post measurement result.

Here, the measurement modules of the reliability measuring units 321 to 323 may vary according to variables used to measure the reliability.

The search control unit 33 operates the search unit and the reliability measurement unit to perform community search. In particular, the search control unit 33 drives the corresponding search unit and the reliability measurement unit according to the search target, and measures each reliability according to the set reliability calculation variable. Optionally drive each negative measurement module. For example, when the search target is a unit, the first search unit 311 and the first reliability measurement unit 321 are driven, and the three variables (community user trust) illustrated above as the variables considered as the first search criteria. When all of the indexes, community activity indexes, and community scales are set, all measurement modules 321a to 321c of the first reliability measurer 321 are driven. However, when only one of the three variables is set as the evaluation variable, only the corresponding measurement module is driven to perform the reliability measurement. This selective driving applies equally to all searching and reliability measuring units.

In this way, the search target and the reliability information for each search target are transmitted to the search result providing unit 34.

The search result providing unit 34 arranges the search targets provided from the search control unit 33 according to the reliability and provides them to the user. As shown in FIG. 3, the search result providing unit 34 includes an arrangement order determination module 341 and a result providing module 342.

The arrangement order determination module 341 may determine the first order information corresponding to the order determined when providing the search results in the system 100, the second order information corresponding to the order determined by the user's selection, or the characteristics of the search word. At least one order information of the third order information corresponding to the order is provided to the result providing module 342. The result providing module 342 arranges the search results for each search object according to the provided order information, generates a search result page, and provides the search result page to the user terminal 300.

Therefore, the search results may be provided on the search result page, for example, in the order of the unit, the post, and the sort menu, or the search results may be provided in the order of the post, the unit and the sort menu.

In this case, according to an embodiment of the present invention, the search results are arranged in order of high reliability for each search target, or only the search results having a reliability higher than or equal to a predetermined value are provided.

Meanwhile, in the embodiment of the present invention, the first and second community servers 21 and 22 are implemented in one system, but may be implemented separately in different systems. For example, the first system may include a first community server to provide a public community service, and the second system may include a second community server to provide a personal community service. In addition, in the above embodiment, the database and the processing units 211 and 221 in each server 21 and 22 may be implemented as independent servers to process corresponding functions, and each database may be integrated into one server. Incidentally, in the present invention, the database is classified as described above for convenience of description, but is not limited thereto.

Next, a search method through a network according to an embodiment of the present invention will be described based on such a structure.

5 is a flowchart illustrating a search method according to an embodiment of the present invention.

As shown in FIG. 5, a plurality of users directly access the search server 30 by executing a user program provided by the interface server 10 in order to search for desired content using the terminal 300. Or directly access the interface server 10 through a web browser and input a search word corresponding to the content to be searched. In this case, the user may access the initial web page of the system through the interface server 10, access a predetermined web page providing a community service, and then input a search word on the community service web page. For example, after accessing the system 100, a user may input a search word by accessing a page providing a public community service or input a search word by accessing a page providing a personal community service (S100 to S110).

The search word input by the user is transferred to the search server 30, and the search server 30 performs a community search based on the input search word.

First, the search server 20 searches for information including the input search word for each search target (S120).

In detail, the search server 20 searches for a unit having information including the search word among the units, that is, a community. Search the content databases 214, 223 of the first and / or second community servers 21, 22 to search for communities with content that includes information that matches the entered search terms, or each community server 21, The registration database 212, 222 of 22) is searched for a community having a description including a word matching the search word among the description of the community (S130 to S140).

Next, the reliability of the searched communities is measured according to the evaluation variable set to be considered in the first search criterion. That is, the reliability of the community is calculated in consideration of at least one or more evaluation variables among community user confidence index, community activity index, and community size (S150).

In addition, the search server 20 finds a post having information including the search word in the post. That is, among the postings stored in the content databases 214 and 223 of the first and second community servers 21 and 22, a posting including information corresponding to the input search word is searched for (S120). In general, the posts posted in the community are composed of a title and content, and the search server 30 determines whether information corresponding to the search word is included by referring to the title and content of the post (S160). S170).

Next, the reliability of the searched posts is measured according to the evaluation variable set to be considered in the second search criterion. That is, the reliability of the article is calculated in consideration of at least one or more evaluation variables among the evaluation variables such as the user's characteristics, content usage, additional information attached, content size, etc. (S180).

In addition, the search server 20 searches for a classification menu having information including the search word among the classification menus of each community (S190). That is, among the postings stored in the content databases 214 and 223 of the first and second community servers 21 and 22, find a post including information matching the input search word, and check the categorization menu displaying the posting. do.

Next, the reliability of the posts of the searched classification menu is measured according to the evaluation variable set to be considered in the third search criterion. That is, the reliability of the classification menu is calculated in consideration of at least one or more evaluation variables among the evaluation variables such as the number of posts (S200).

When the reliability is calculated as described above, the searched community is arranged and provided to the user according to the calculated reliability (S210 to S220). At this time, the search server 30 provides each search target according to the set search arrangement order. Arrange and provide search results for each search target. For example, in the system, if the sort order is set in the order of units, posts, and sorting menus, the search results page displays the search results for the unit first, followed by the search results for the posts, and finally the sorting menu. The search results are displayed. The order of these search results can be changed. On the other hand, the search results may be displayed in order according to the measured reliability, or randomly arranged regardless of the reliability, or only those having a certain reliability or more than a predetermined reliability may be selected and displayed.

In the searching process of the present invention performed as described above, the searching steps S120 to S200 will be described in more detail.

6 is a flowchart illustrating a process of searching for a unit in a search step in more detail.

As shown in FIG. 6, when a search word is provided, the search control unit 33 of the search server 30 first drives the first search unit 311 to search for communities including information corresponding to the search word. . As described above, the first search unit 311 searches the content databases 214 and 223 of each community server 21 and 22 to find communities having contents including the search word. Or search the registration database (212, 222) of each community server (21, 22) to find a community having a community description or name including the search word (S300 ~ S320). At this time, the community may be selectively searched according to the word frequency indicating how many words matched with the search word or the degree of matching with the search word.

When a community having information corresponding to the entered search word is searched, the search control unit 33 selectively selects the measurement modules 321a to 321c of the first reliability measuring unit 321 according to the set evaluation variable. Drive it. For example, when only the community user confidence index is set as an evaluation variable to be considered when measuring reliability, only the reliability index measurement module 321a is driven. When only the community activity index is set as the evaluation variable, the activity index measurement module 321b is used. Drive only). In addition, when two or more evaluation variables are set, each of two or more measurement modules is driven. When two or more evaluation variables are set as described above, the measurement module is simultaneously driven to perform the measurement, or the measurement module is sequentially driven to perform the sequential measurement.

Each measurement module 321a to 323c of the first reliability measurer 32 obtains a value corresponding to the evaluation variable.

Specifically, when the searched community is a public community, the confidence index measurement module 321a searches the member database 213 of the first community server 21 based on the ID of the public community, and registers as a member in the public community. After obtaining the confidence indices of each user, the community confidence index is calculated by summing the confidence indices of the obtained users. On the other hand, if the searched community is a personal community, the registration database 222 of the second community server 22 is searched to find a trust index of the user who established the personal community and set as the community user trust index (S330 to S). S350).

The activity index measurement module 321b obtains the number of recent contents, the number of recent visitors, the number of relationship formation, and the like for each searched community, and then measures the activity index based on the obtained values. For example, if the searched community is a public community, search the registration database 212 of the first community server 21 to obtain the number of recent visitors based on the ID of the public community, and search the member database 213 for community members. The activity index is calculated, and then the community activity index is calculated by calculating the sum of the activity indexes of the members and adding the recent visitor numbers. In addition, when the searched community is a personal community, the registration database 222 of the second community server 22 is searched to obtain the number of recent visitors and relationship formation, and the content index 223 is searched for the activity index of the corresponding community operator. Then, the community activity index is calculated by calculating the calculated activity index, the number of recent visitors, the number of relationship formation, and the like (S360 to S380).

The scale measurement module 321c searches the registered databases 212 and 222 and the contents database 214 and 223 of each community for each searched community, obtains the total number of members and the total number of contents of the corresponding community, and then scales the community based on the obtained values. Is calculated (S390 to S410).

As described above, the values measured in the respective measurement modules 321a to 321c of the first reliability measurer 321 are provided to the reliability calculation module 321d, and the reliability calculation module 321d corresponds to the set evaluation variable. The reliability of the searched community is calculated based on the values (S420). That is, the reliability of the community is calculated by considering the measurement values corresponding to at least one of community user confidence index, community activity index, and community size according to three evaluation variables. For example, the community reliability is calculated through the method of summing all values corresponding to the community user confidence index, community activity index, and community size.

As described above, the information about the community searched by the first search unit 311 and the reliability of each community measured by the first reliability measurer 321 are transmitted to the search result provider 34.

The result providing module 342 of the search result providing unit 34 selects only the communities having a predetermined value or more based on the reliability of each searched community, and then generates a search result page and provides the result to the user, or The community is arranged in a high order to generate a search result page and then provided to the user (S430). Alternatively, each searched community may be randomly arranged to generate a search result page. On the search results page, link information (eg, community homepage, etc.) that provides direct access to the community is displayed in association with the community's name. In addition, the description of the community, moderator nickname (or ID), number of members, creation date, Information such as a directory to which a community belongs may be displayed.

When searching for the community described above, if content having information corresponding to the input search word is searched, in addition to providing information about the community as shown in the above-described embodiment, the search result for the searched content is displayed. Can provide. That is, link information that can be directly linked to the searched content on the search result page may be displayed based on a content title, a part of the content for each content, and a community name where the content is posted. In this case, the content search result may also be arranged according to the reliability of each community calculated according to the above-described embodiment, or only the content posted in the community whose community reliability is greater than or equal to a set value may be selected and displayed.

Next, a process of searching for a post in the search step shown in FIG. 5 will be described in detail.

7 shows a post search process in more detail.

As shown in FIG. 7, when a search word is provided, the search controller 33 of the search server 30 drives the second search unit 312 to search for contents including information corresponding to the search word. The second search unit 312 searches the content databases 214 and 223 of each community server 21 and 22, finds the contents including the search word, and transfers the contents to the reliability measurer 32 (S500 to S510). At this time, the content may be selectively searched according to the word frequency indicating how many words matching the search word are included or the degree of matching with the search word.

Next, the search controller 33 selectively drives the measurement modules 322a to 322d of the second reliability measurer 322 according to the set evaluation variable (S320). For example, when only the user characteristic is set as an evaluation variable to be considered in search, only the characteristic measurement module 322a is driven, and when the content usage is set as the evaluation variable, only the utilization measurement module 322b is driven. In addition, when the additional information is set as the evaluation variable, only the additional information measuring module 322c is driven. When the content size is set as the evaluation variable, only the size measuring module 322d is driven. In addition, when two or more evaluation variables are set, each of two or more corresponding modules is driven. When two or more evaluation variables are set as described above, a search is performed by simultaneously driving a corresponding module or a sequential search is performed by driving a module sequentially.

Each measurement module 322a to 322d of the second reliability measurer 322 obtains a measurement value corresponding to the evaluation variable.

In detail, the characteristic measurement module 322a may use the member database 213 based on the IDs of users who have posted the contents searched by the second search unit 312 when the characteristics of the user are set as the evaluation variables and the age is set among the user characteristics. And / or search the registration database 222 to measure the age of the users (S530 to S550). In addition, if a job is set among the user characteristics, as described above, the member and / or registration databases 213 and 222 may be searched to determine the job of the user.

In addition, when a user confidence index is set among user characteristics, the content posted by a predetermined user is found in the content databases 214 and 223 of each community server 21 and 22, and the user's confidence index is measured based on the number of scraps of each content. can do. Alternatively, the user's trust index may be measured based on the number of other users, that is, the number of neighbors, formed from the registration databases 212 and 222 of each community server and forming a predetermined relationship with the community to which the user has joined or opened. In addition, the user's confidence index can be measured in various ways.

On the other hand, when the utilization of the content is set as an evaluation variable, the utilization measurement module 322b refers to the number of replies and / or comments and / or e-mails corresponding to each of the retrieved contents with reference to the content databases 214 and 223. The utilization is measured based on the number of times sent (S560 to S580).

In addition, when it is determined whether additional information is attached as an evaluation variable, the additional information measuring module 322c searches the content databases 214 and 223 of each community server 21 and 22 to find contents including the search word, It is checked whether the additional information such as an image, a video, audio, etc. is attached to the content other than the text information, and the type of the attached additional information is determined (S590 to S610).

In addition, when the size of the content is set as an evaluation variable, the size measurement module 322d searches the content databases 214 and 223 of each community server 21 and 22 to find contents including the search word, and the size of the found contents. (Byte count, etc.) are measured (S620 to S640).

As described above, the values measured in each of the measurement modules 322a to 322d of the second reliability measurer 322 are provided to the reliability calculation module 322e, and the reliability calculation module 322e corresponds to the set evaluation variable. The reliability of the retrieved content is calculated based on the measured values (S650). That is, the reliability of the content is calculated based on measurement values corresponding to at least one evaluation variable among evaluation variables such as a user's characteristics, content usage, additional information attached, content size, and the like.

Various methods may be used to calculate the reliability. In the present embodiment, a reliability generation function is set, and the reliability generation function calculates the reliability of the content based on a weight according to a predetermined input value according to the measured value of each evaluation variable. Here, the input value input to the reliability generation function may be a measurement value for each evaluation variable or may be different.

In particular, according to an embodiment of the present invention, a plurality of setting ranges having representative values are provided for each evaluation variable, and the representative value of the setting range to which the measured value for each evaluation variable belongs is set as an input value of the reliability generation function. For example, if the evaluation variable is a user characteristic and is specifically set to age, the representative value is "10" in the case of a teenager, the representative value is "50" in the case of a teenager, and in the 30s. The representative value can be given as "100". In this case, when the measured age of the user belongs to the twenties, the input value is set to "50".

On the other hand, if the evaluation variable is the content size, the representative value may be proportionally given among the contents corresponding to the maximum limit while the size of the content is greater than or equal to the set value. For example, if 500 bytes is set as the maximum limit of the size of content that can be created with content, the maximum representative value is given when the size of the predetermined content is 500 bytes, and the representative value is lower when it is larger than 500 bytes. Grant. This takes into account that too much content may contain more unnecessary information, but is not necessarily limited to this method.

The weight is based on the input values for each evaluation variable, and the reliability of the corresponding content is calculated based on the reliability generation function. However, the method of calculating reliability according to the present invention is not limited thereto. For example, other methods, such as calculating reliability by summing all the measured values for each evaluation variable, may also be used.

Next, the reliability of each content measured by the second reliability measurer 322 as described above together with the content retrieved by the second searcher 312 is transmitted to the search result provider 34.

The search result providing unit 34 selects only the content having a predetermined value or more based on the reliability of each searched content and generates a search result page based on this and provides the search result page to the user, or arranges the contents in the order of high reliability. The search result page is generated and then provided to the user (S660). Alternatively, the searched content page may be generated by arranging the searched content randomly.

On the search results page, the title, part of the content, author, creation date, source, etc. of the content may be displayed. In addition, if a file such as an image is attached to the content, a separate icon may be used to indicate whether to attach the file. I can display it. Alternatively, the reliability calculated as described above for each content may be displayed. Therefore, the user can easily select and use the most reliable contents including the search word.

Meanwhile, even in the process of searching the classification menu in the search step shown in FIG. 5, similar to the above unit and the posting search process, the classification menus providing information on the search word are found, and the reliability classification is found for the found classification menu. Is performed. That is, under the control of the search controller 33, the third search unit 313 is driven to match the input search word among postings stored in the content databases 214 and 223 of the first and second community servers 21 and 22. Find the post that contains the information you are looking for, and find the category menu that contains the post.

Next, the post measurement module 323a of the third reliability measurer 323 checks the number of posts posted in each classified menu based on the content databases 214 and 223, and measures the reliability of each classified menu based on this. do. The search result providing unit 34 displays the classified menus searched according to the reliability and provides the search to the user.

Meanwhile, the method of calculating the reliability of the searched post in the above-described post search process using the representative value, the input value, and the reliability generation function may be equally applied to the unit and classification menu searches. A person skilled in the art may devise a method for calculating a reliability of a community and a reliability of a classification menu using a representative value, an input value, and a reliability generation function when searching for a unit based on the method illustrated in FIG.

Next, as described above, another embodiment in which the search is performed according to different search criteria for each search target and the search results given the reliability are arranged.

In the present exemplary embodiment, the searched units, the posts, and the classification menus are not arranged in the order set in the system, and the system automatically changes the order according to the input search word to arrange the search results.

8 illustrates a structure of an additional configuration of a search server according to another embodiment of the present invention.

As illustrated in FIG. 8, the search server 30 according to an embodiment of the present invention may include a search log processing unit 35 for generating and recording search log information of users after providing a search result, and a search log database in which log information is recorded. 36), and further, a search log analysis unit 37 which analyzes a search target selected by the user most by each search term based on the log information stored in the search log database 36 and stores the result corresponding to the search term. Include.

The search log processing unit 35 analyzes which object the user selects on the search result page and records the result. Specifically, the search result for each search object such as a unit, a post, and a classification menu provided on the search result page is recorded. The search results of the search results are selected from among the user, and the identified information is matched with the input search word and stored in the search log database 36. Accordingly, the search log database 36 records information about the search target that the user first selected on the search result page corresponding to the search word, that is, search log information. For example, the search target "unit" selected in response to the search word "car" is recorded, and in addition, the selected user ID is recorded.

Thereafter, the search log analysis unit 37 analyzes the search object most selected by the users based on the search log information of the users and stores the result corresponding to the search word. Therefore, the search log analysis result is stored in the search log database 36 corresponding to the search word. The search log analysis result shows the search targets most selected by the users in order, and, for example, order information such as a unit, a post, and a classification menu. The log analysis result may be different for each search word.

9 illustrates a search result providing method according to another embodiment of the present invention in detail.

As described above, when a search is performed according to different search criteria for each search target and reliability is granted, the arrangement order determining module 341 of the search result providing unit 34 performs a search log database based on the input search word. (36) is searched and the log analysis result corresponding to the search word is referred to (S700 to S710).

In addition, the arrangement order of providing the search result is determined according to the log analysis result (S720). For example, let's say the search term entered is "car." Until now, when a search is performed on the search word and the search result is provided to the user, the search targets that the user first selects and check the search result have the most units such as a cafe or a blog, and then, the classification menu. And, if the post was made in order, the log analysis result for this order is stored in the search log database 36. Accordingly, the arrangement order determination module 341 determines the arrangement order in order of units, classification menus, and posts according to the log analysis result, and provides the result to the result providing module 342.

Once the arrangement order is determined, the result providing module 342 arranges the search objects in the order of units, classification menus, and posts, and the search results for each search object are based on the reliability measured by different search criteria as described above. Arranged accordingly, a search result page is generated (S730).

Meanwhile, in order to check the search results in more various states, the arrangement order determining module 341 of the search result providing unit 34 provides a menu for changing the order. Accordingly, when the user requests a search result arrangement order change (S740), the arrangement order is determined and provided to the result providing module 342 according to the request, and the result providing module 342 according to the arrangement order selected by the user Rearrange the search targets to present the search results to the user. In other words, if the user requests to change to post, sort menu, and unit in the order of unit, sort menu, and post, the search target is displayed on the search result page in order of post, sort menu, and unit. The search results for the object are arranged and provided according to the reliability (S750).

As the search result page generated as described above is provided to the user, the user may check the search results in the order of search targets that are mainly checked on the search result page.

On the other hand, if a user clicks on a search result of a predetermined search target and selects the search result on the search result page provided as described above, the search log processing unit 35 checks the search target selected first by the user and searches log information accordingly. Stored in the log database 36 (S760 to S780). For example, a search target, for example, that is first identified in response to a search word is stored, and a unit is additionally recorded.

According to the recording of the search log information, the log analysis of the search log analysis unit 37 is performed, so that the tendency of checking the search results of the users may be reflected in the search.

Meanwhile, the log analysis result may be recorded by generating and analyzing search log information for all search terms, but in this case, since the data throughput is too high and the search terms entered may not have been previously entered and processed, Search log information can be generated for each category and analyzed. For example, entertain search terms. Divide categories such as society, culture, and politics (it is also possible to divide multiple subcategories for each category), and find and record log information by finding the search targets that users first identify as described above. Analyzed log information. Afterwards, the search term inputted during the search belongs to a category, and the search order may be determined according to the log analysis result of the corresponding category to provide the search result to the user.

Meanwhile, the above-described embodiment divides the search target into a unit, a post, and a classification menu. However, the search target according to the present invention is not limited thereto, and all targets that can be searched on a network such as web documents, news, images, etc. on the web are available. This may be included. The search criteria may be set differently for each of these search targets, and the search criteria may use at least one or more predetermined evaluation variables according to the first to third search criteria described above, but is not limited thereto.

The retrieval method described above may be implemented in the form of a program stored in a computer-readable recording medium. The recording medium may include any kind of recording device that stores data that can be read by a computer. For example, a CD-ROM, a magnetic tape, a floppy disk, and the like may be used. Included is implemented in the form of).

Although the present invention has been described with reference to the most practical and preferred embodiments, the present invention is not limited to the above disclosed embodiments, but also includes various modifications and equivalents within the scope of the following claims.

As described above, according to an embodiment of the present invention, the search is performed according to different search criteria for each search target, thereby providing more reliable information for each search target.

In addition, the search results are arranged according to the reliability, so that the user can use the reliable content more easily.

In addition, since the search result arrangement order can be changed, users can arrange and search the search results according to their own preferences.

In addition, the results for each search target are provided in order of search targets checked by the users, according to the search word, so that the users can easily find a search target that provides more reliable information.

Therefore, the user is provided with reliable information more conveniently.

1 is a structural diagram of a search system through a network according to an embodiment of the present invention.

2 to 4 are detailed structural diagrams of each server constituting the search system shown in FIG. 1.

5 is a flowchart illustrating a search method according to an embodiment of the present invention.

FIG. 6 is a detailed flowchart of a unit searching method in the searching step shown in FIG. 5.

FIG. 7 is a detailed flowchart illustrating a post search method among the search steps shown in FIG. 5.

8 is a structural diagram of a search server according to another embodiment of the present invention.

9 is a flowchart illustrating a search result providing method according to another exemplary embodiment of the present invention.

Claims (1)

In a method of searching for information on a search word provided from the user terminal connected to a plurality of user terminals through a network, a) setting a search target to perform a search based on the search word if a search word is provided from a user terminal through the network; b) searching for information including the search word for each search target; c) calculating reliability based on different search criteria with respect to the information searched for each search target; d) determining an order in which search targets are arranged according to the search word; And e) arranging search objects according to the arrangement order, and arranging searched information for each search object according to the measured reliability and providing the user with the measured reliability. Search method comprising a.