KR20100138858A - Method and system for providing relevance site using chracteristic word of site - Google Patents

Abstract

PURPOSE: A method and a system for providing a relevance site using a characteristic word of a site are provided to expose search words for access to various sites, thereby providing information for the search words when accessing the sites. CONSTITUTION: A search word exposing unit(601) exposes one or more search words inserted into a site. If one of the exposed search words is selected, an associated site detecting unit(602) detects an associated site with the search word. A characteristic word determining unit(604) collects sites. The characteristic word determining unit determines at least one specific word about the collected sites. If the search word is included in the determined specific words, an associated site determining unit(605) determines the search word and the site as an associated site.

Description

METHOD AND SYSTEM FOR PROVIDING RELEVANCE SITE USING CHRACTERISTIC WORD OF SITE}

The present invention relates to a method and system for providing a related site, and more particularly, to a related site providing method and system for providing a site associated with a search word and a site associated with a specific site.

Users often want to know what search terms they access through sites like their own homepages, blogs, or cafes. For example, a user who posts articles related to "wine" or "soccer" on a blog may want to know what search terms others are accessing. By analyzing the incoming search words, it is possible to grasp the interests of others, and the user may have more opportunities to more effectively expose his or her posts by reflecting the interests of others. In addition, by analyzing the searched words to find out the interests of others, you can more effectively advertise and expose your blog.

In addition, by analyzing the imported search words, a need to search for a site related to the search word may be required. Furthermore, the need to search for relevant sites for a particular site may be raised. By backtracking these search terms, you can effectively promote your site. In addition, the contents of the site's documents can be posted more effectively to the public. As a result, there is a need for a satisfactory invention that requires such potential promotional effects.

The present invention provides a method and system for providing a related site that can provide information about a search word introduced in accessing the site by exposing a search word introduced to access various sites such as a mini homepage, a blog or a cafe.

The present invention provides a related site providing method and system that can provide additional information on a search word by providing a related site linked with the search word when selecting the exposed search word.

The present invention provides a method and system for providing a related site that can easily search for related sites that reflect the content of the specific word by detecting the related site by using characteristic words of the site such as search terms, index words, tag words, and category words. do.

The present invention provides a method and system for providing a related site that can detect a related site for a specific site by grasping the degree of association between the sites by using characteristic information between sites such as a mini homepage, a blog or a cafe.

In accordance with another aspect of the present invention, there is provided a method for providing a related site, including: exposing at least one search term introduced to a site; detecting one of the at least one search term related to the search term; And providing the detected related site by linking with the selected search word.

In accordance with an embodiment of the present invention, a method for providing an associated site includes determining at least one characteristic word for a site and generating a word vector of each characteristic word, and using the generated site word vector, a word vector between sites. Calculating an association degree and detecting an association site using the word vector association degree.

Related site providing system according to an embodiment of the present invention is a search term exposure unit that exposes at least one search term introduced to the site, when any one of the at least one of the exposed search terms, the associated site for the search term is detected And an associated site detector for connecting the selected search word with a link and providing the detected related site.

According to an embodiment of the present invention, an associated site providing system includes a word vector generator configured to determine at least one feature word for a site and to generate a word vector of each feature word, and between sites using the generated site word vector. And a word vector relevance calculator for calculating a word vector relevance of and a related site detector for detecting a related site using the word vector relevance.

According to the present invention, a method and system for providing a related site that can provide information on a search word introduced to access the site by exposing a search word introduced to access various sites such as a mini homepage, a blog or a cafe, etc. do.

According to the present invention, a method and system for providing a related site which can provide additional information about a search word by providing a related site linked with the search word when selecting the exposed search word are provided.

According to the present invention, a method and system for providing a related site that can easily search for a related site reflecting the content of the specific word by detecting the related site by using the characteristic word of the site such as a search word, an index word, a tag word, and a category word. This is provided.

According to the present invention, there is provided a related site providing method and system capable of detecting a related site for a specific site by grasping the degree of association between sites using characteristic information between sites such as a mini homepage, a blog or a cafe.

1 is a flowchart illustrating a method for providing a related site according to an embodiment of the present invention.
2 is a diagram illustrating an example of exposing a search word introduced to a site according to a related site providing method according to an exemplary embodiment of the present invention.
3 is a diagram illustrating a process of providing a related site for a selected search word according to an embodiment of the present invention.
4 is a flowchart illustrating a method for providing a related site according to another embodiment of the present invention.
FIG. 5 is a diagram for describing a process of detecting an associated site through a degree of association between a site and a site word vector between sites according to an embodiment of the present invention.
6 is a block diagram illustrating an associated site providing system according to an embodiment of the present invention.
7 is a block diagram illustrating a related site providing system according to another exemplary embodiment of the present invention.

Hereinafter, with reference to the contents described in the accompanying drawings will be described in detail an embodiment according to the present invention. However, the present invention is not limited to or limited by the embodiments. Like reference numerals in the drawings denote like elements. The method for providing a related site according to an embodiment of the present invention may be performed by a related site providing system.

1 is a flowchart illustrating a method for providing a related site according to an embodiment of the present invention.

The method of providing a related site according to an embodiment of the present invention may expose at least one search word introduced to a site (S101). For example, in the step S101 of exposing at least one search word introduced to the site, the site inflow rate for each of the at least one search word and the at least one search word may be exposed. In this case, in the step of exposing at least one search word introduced to the site (S101), the at least one search word may be arranged and exposed according to the inflow rate of each search word. In one example, the site may include a minihompy, a blog or a cafe. The type of site is not limited, and any site can be included as long as it can be accessed through a link such as a URL online.

For example, the site inflow rate for a search term may be calculated by the number of times of accessing the corresponding site and the number of accesses per search term. For example, it can be assumed that a search term 'a', 'b', or 'c' enters a site called 'A'. In this case, the total number of accesses to 'A' site was 100, of which 50 flowed into the search term 'a', 37 flowed into the search term 'b', and flowed into the search term 'c'. Say 13 times. Then, the site inflow rate for each of the search terms may be calculated as 'a' 50%, 'b' 37%, and 'c' 13%. In operation S101, at least one search word introduced to a site may be exposed by sorting in order of 'A', 'b', and 'c' according to the site inflow rate.

In the method of providing a related site according to an embodiment of the present invention, when any one of the at least one search term is selected, the related site for the search term may be detected (S102). Referring to the above example, when the user selects the 'b' search word, detecting the related site for the search word (S102) may detect the related site for the 'b' search word.

More specifically, if the search query 'Jazz' is the highest search inflow rate for the blog 'abcde' and the user selects 'Jazz', then the relevant site is provided by the blog, cafe or Related sites such as minihompy can be detected and provided. That is, the method for providing a related site according to an embodiment of the present invention may provide a related site associated with a specific search word.

In one example, detecting the associated site for the search word (S102) may collect the site to determine at least one characteristic word for the collected site. At this time, the step (S102) of detecting a related site for a search word may collect all the ranges of sites that can be collected and store them in a database. In operation S102, detecting an associated site for a search word may determine at least one characteristic word for a site stored in a database. In this case, the characteristic word may be a word representing a representative characteristic of the corresponding site. However, when various kinds of articles are posted through a plurality of lists in the site, the characteristic word may be one or more.

In this case, the determining of the at least one characteristic word may determine the at least one search word introduced to access the document of the site as the characteristic word of the site. For example, the at least one search word introduced to access the document of the site may be determined through the search log data accumulated during the preset period.

In the determining of the at least one characteristic word, the at least one index word for the document included in the site may be determined as the characteristic word of the site. That is, the index word that can represent the characteristics of the documents included in the collected site may be determined as the characteristic words of the site. For example, in the case of a blog related to 'wine', a document included in the blog may include content related to 'wine' and an index word may be a word related to 'wine'.

At this time, when a word that is frequently exposed or a word that is rarely exposed to the entire document of the site becomes an index word, the index word may not efficiently reflect the characteristics of the site. Accordingly, in the determining of the at least one characteristic word, the index word belonging to a preset document exposure frequency range among the at least one index word may be determined as the characteristic word of the site.

In the determining of the at least one characteristic word, the at least one tag word input when a document included in the site may be determined as the characteristic word of the site. In this case, the tag word may be a word input by the document creator as a representative feature of the content of the document.

In the determining of the at least one characteristic word, the at least one category word for classifying a document included in the site may be determined as the characteristic word of the site. That is, a category (eg, music, travel, photo, etc.) generated by the administrator of the site to classify the document may be a characteristic word of the site.

In operation S102, when the determined at least one characteristic word is related to the search word, the site may be determined as the related site for the search word.

For example, if the search term "A" is selected, the site with the attribute word determined of the collected sites associated with "A" contains documents related to the search term "A". There is a connection.

In this case, when the search word and the characteristic word are not limited to the same thing and are related in the same or similar fields, the site including the characteristic word may be detected as the related site for the search word.

The method for providing a related site according to an embodiment of the present invention may provide the detected related site by linking the selected search word with a link (S103). In one example, providing a detected related site (S103) may provide a list of a plurality of related sites that are associated with a search term. Also, as an example, providing a detected related site (S103) may provide a document in which the selected search word is exposed among documents included in the detected related site.

2 is a diagram illustrating an example of exposing a search word introduced to a site according to a related site providing method according to an exemplary embodiment of the present invention.

FIG. 2 illustrates a search result 201 that provides a ranking of search terms introduced to a site 'A' when a user inputs a site 202 'A'. As mentioned earlier, the site can include a minihompy, a blog or a cafe.

Referring to the search word list 203, at least one search word most introduced among search words introduced to the 'A' site is illustrated. Each of the search terms may be exposed together with an inflow rate, and the search term ranking may be determined according to the inflow rate.

The related site providing method according to an embodiment of the present invention may provide a related site for the selected search word when the user selects the exposed search word. For example, when the user selects the first priority search word 'ab', the related site providing method may detect and provide a related site related to 'ab'. In the same manner, when the user selects the second ranking search term 'cd', the related site providing method may detect and provide a related site related to 'cd'. In this case, a detailed process of providing a related site for the selected search word will be described in detail with reference to FIG. 3.

3 is a diagram illustrating a process of providing a related site for a selected search word according to an embodiment of the present invention.

Referring to FIG. 3, when the user selects the search term 'ab' having the highest ranking, the 'site A' 302, 'site B' 303, 'site C' associated with the term 'ab', 304 can be detected and provided. In one example, the detected sites may be provided in the form of a list. In addition, the detected site may be exposed and provided together with the document including the selected search word 301. In addition, the detected site may be exposed and provided together with a document including a word associated with the selected search word 301.

As already described in FIG. 2, at least one search term entered into the site from the collected site, at least one index word for the document of the site, at least one tag word associated with the site, or at least to classify the document of the site. Any one of the category words may be determined as a characteristic word. The related site providing method may detect the site as the related site when the determined characteristic word and the search word 301 are related to each other.

4 is a flowchart illustrating a method for providing a related site according to another embodiment of the present invention. 4 illustrates a related site providing method for providing a related site related to a specific site, unlike FIG. 1.

In the method of providing a related site according to an embodiment of the present invention, at least one feature word for a site may be determined to generate a word vector of each feature word (S401). For example, in operation S401, the word vector of each feature word may be determined as the feature word of the site, at least one search word introduced to access the document of the site. In this case, the generating of the word vector of each characteristic word (S401) may include generating the word vector for the at least one search word by using the total number of accesses to the document of the site and the access times of each of the at least one search word. Can be generated. That is, when the characteristic word is a search word, the word vector may be calculated as a site inflow rate for the search word.

For example, generating a word vector of each feature word (S401) may determine at least one index word for a document included in the site as the feature word of the site. At this time, when a word that is frequently exposed or a word that is rarely exposed to the entire document of the site becomes an index word, the index word may not efficiently reflect the characteristics of the site. Accordingly, in the generating of the word vector of each feature word (S401), an index word belonging to a preset document exposure frequency range among the at least one index word may be determined as the feature word of the site.

At this time, generating the word vector of each characteristic word (S401) may calculate the word vector by calculating the ratio of each index word to the total index word in the document of the site. For example, the word vector may be calculated in the form of index 1 for 30% and index 2 for 20%.

For example, generating a word vector of each feature word (S401) may determine at least one tag word input when a document included in the site is a feature word of the site.

In this case, in the generating of the word vector of each characteristic word (S401), the word vector may be calculated by calculating a rate at which each tag word is exposed for all the tag words in the document of the site. For example, the word vector may be calculated in the form of tag word 1 of 30% and tag word 2 of 20%.

For example, generating a word vector of each feature word (S401) may determine at least one category word that classifies a document included in the site as the feature word of the site.

In this case, in the generating of the word vector of each characteristic word (S401), the word vector may be calculated by calculating a ratio of the exposure of each category word with respect to all category words in the document of the site. For example, the word vector may be calculated in the form of category word 1 for 25% and category word 2 for 14%. Unlike FIG. 1, the characteristic word in FIG. 3 may be determined for each of the sites to be compared with the sites to be compared.

In the method of providing a related site according to an embodiment of the present invention, a degree of word vector association between sites may be calculated using the generated site word vector (S402). At this time, the step S402 of calculating the word vector association between the sites may calculate the word vector association between the sites using the size of the site word vector. That is, the word vector association may be calculated based on weights between the word vectors.

In one example, calculating the word vector association between sites (S402) may calculate the angle between the site word vector using cosine similarity (cosine similarity). In other words, the cosine similarity can be said to use the dot product between the two vectors.

For example, suppose that the characteristic word for 'site A' is 'wine, music and travel to Europe' and the characteristic word for 'site B' is 'wine, photo and travel to France'. The word vector for each characteristic word may be (0.6, 0.3, 0.1) for 'A site' and (0.3, 0.5, 0.2) for 'B site'. Then, the word vector association may be determined as 0.6 * 0.3 + 0.1 * 0.2 = 0.2.

For example, when a high degree of association between characteristic words of a site is applied, a method of weighting each characteristic word may also be applied. That is, 'site A' and 'site B' have the same characteristic words as 'wine', and thus have a higher degree of relevance and thus can be given more weight than other characteristic words.

In a method of providing a related site according to an embodiment of the present invention, an associated site may be detected using the word vector degree of association (S403). For example, the detecting of the related site (S403) may include determining a distance between the sites according to the word vector relevance and sorting the sites in the order of the shortest distance between the sites to determine which sites belong to a preset ranking. Detecting as an associated site.

For example, if the word vector association is 0.7, the site B and the word vector association is 0.7, and the site C and the word vector association is 0.4. It can be judged that the distance from the 'site B' is closer than the 'site C'. Detecting the related site (S403) is to arrange the sites in the order of short distance to the site A as a reference, and to detect and provide only sites belonging to a predetermined rank (for example, 5th) as the related site. Can be.

FIG. 5 is a diagram for describing a process of detecting an associated site through a degree of association between a site and a site word vector between sites according to an embodiment of the present invention.

Referring to FIG. 5, the method for providing a related site according to an embodiment of the present invention may detect and provide a related site for 'A site 501' which is a reference. At this time, the related site providing method may detect the related site by determining the degree of association with the collected sites 'B site 502', 'C site 503' and 'D site 504'. At this time, the related site providing method is at least one from each of the 'site A (501), the reference site' B site 502 ',' C site 503 'and' D site 504 'to be compared The characteristic word of can be determined. As already mentioned above, feature words can be determined through search terms, index words, tag words, or category words for each site. In addition, the method of providing a related site may determine a word vector for each of the determined characteristic words.

In this case, the method of providing a related site is a word vector association between a word vector of 'A site 501' and a word vector of each of 'B site 502', 'C site 503' and 'D site 504'. The degree can be calculated. In addition, the method of providing a related site may determine that the distance between the sites is closer as the word vector association is greater. Also, for example, it is assumed that the related site providing method is arranged in the order of the distance closest to the 'site A (501)', 'site C (503),' site D (504) 'and' site B (502) '. lets do it. In addition, suppose that the related site providing method detects the top 2 positions as the related site. Then, the related site providing method may detect and provide 'C site 503' and 'D site 504' belonging to the top two as related sites.

6 is a block diagram illustrating an associated site providing system according to an embodiment of the present invention. Related site providing system 600 according to an embodiment of the present invention may include a search word exposure unit 601, an associated site detection unit 602 and an associated site providing unit 603.

The search word exposure unit 601 may expose at least one search word introduced to the site.

In this case, the search word exposure unit 601 may expose a site inflow rate for each of the at least one search word and the at least one search word. In addition, the search word exposure unit 601 may sort and display the search terms according to the site inflow rate for each search word.

The related site detector 602 may detect a related site for the search word when any one of the at least one search term is selected. For example, the associated site detector may include a feature word determiner 604 and an associated site determiner 605.

The feature word determiner 604 may collect sites to determine at least one feature word for the collected sites.

For example, the feature word determiner 604 may determine at least one search word introduced to access the document of the site as the feature word of the site.

For example, the feature word determiner 604 may determine at least one index word for a document included in the site as the feature word of the site.

For example, the characteristic word determiner 604 may determine at least one tag input when creating a document included in the site as the characteristic word of the site.

For example, the characteristic word determiner 604 may determine at least one category word for classifying a document included in the site as the characteristic word of the site.

The related site determining unit 605 may determine the search word and the site as the related site when the search word is included in the at least one characteristic word.

The related site provider 603 may provide the detected related site by linking the selected search word. At this time, the related site provider 603 may provide a list of related sites. In addition, the related site providing unit 603 may provide a document in which the search word is exposed among documents included in the detected related site.

7 is a block diagram illustrating a related site providing system according to another exemplary embodiment of the present invention.

The associated site providing system 700 according to an exemplary embodiment of the present invention may include a word vector generator 701, a word vector association calculator 702, and an associated site detector 703.

The word vector generator 701 may generate at least one feature word for the site to generate a word vector of each feature word.

For example, the word vector generator 701 may determine at least one search word that is introduced to access a document of a site as a characteristic word of the site. In this case, the word vector generator 701 may calculate a site inflow rate using the total number of accesses to the document of the site and the number of accesses of each of the at least one search word. In this case, the calculated site inflow rate may be a word vector for a search word.

For example, the word vector generator 701 may determine at least one index word for a document included in the site as a characteristic word of the site. In this case, the word vector generator 701 may generate a word vector by calculating a ratio of each index word exposed to all index words in the document of the site.

For example, the word vector generator 701 may determine at least one tag input when the document included in the site is a characteristic word of the site. In this case, the word vector generator 701 may generate a word vector by calculating a ratio of each tag word to the entire tag words in the document of the site.

For example, the word vector generator 701 may determine at least one category word for classifying a document included in the site as a characteristic word of the site. In this case, the word vector generator 701 may generate a word vector by calculating a ratio of each category word exposed to all category words in the document of the site.

The word vector association calculator 702 may calculate a word vector association between sites using the generated site word vector. The word vector association calculator 702 may calculate a word vector association between the sites using the size of the site word vector.

The related site detector 703 may detect the related site by using the word vector degree of association. For example, the related site detector 703 may determine the distance between the sites according to the word vector degree of association. In addition, the related site detector 703 may arrange the sites in the order of short distance between the sites and detect a site belonging to a predetermined rank as the related site.

6 and 7 may refer to the descriptions of FIGS. 1 to 5.

In addition, the method for providing a related site according to an embodiment of the present invention includes a computer readable medium including program instructions for performing various computer-implemented operations. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. The media may be program instructions that are specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, the spirit of the present invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications thereof will belong to the scope of the present invention.

600: related site providing system
601: Query exposed portion
602: associated site detection unit
603: related site provider
604: characteristic word determination unit
605: association site determination unit

Claims (1)

Exposing at least one search term introduced to the site;
If any one of the at least one search term is selected, detecting a related site for the search term; And
Linking with the selected search word to provide the detected related site
How to include.

Images