KR20100013358A - Method and system for providing search service using synonym filtering - Google Patents

Abstract

PURPOSE: A method and a system for providing a search service using synonym filtering are provided to use collocation information of queries determined by a synonym, thereby easily and accurately filtering synonyms. CONSTITUTION: A collocation information extractor(120) extracts collocation information of queries belonging to a synonym group. A filtering unit(140) distinguishes relation of queries belonging to the synonym group. The filtering unit filters the queries belonging to the synonym group according to the distinguished relation.

Description

Method and System for Providing Search Service Using Synonym Filtering}

The present invention relates to Internet search, and more particularly, to a method of providing a search service using synonym filtering.

Due to the development and spread of the Internet, various services using the Internet are provided, and a representative example thereof is a search service. When the search service inputs a word or a combination of words to be searched by a user as a query, the search engine corresponds to a search result document corresponding to the input query (for example, a web site, an article, or a search query including a search query input from the user). An image having a file name including the corresponding search query).

In using such a search service, since it is not easy to select an appropriate query word corresponding to a desired search result, users often input a query word other than the query word corresponding to the search result that they want to obtain. Occurs. In this case, since the search system performs a search based only on the actually entered query, there is a problem that users do not get the search results they want.

In order to solve this problem, the search system may also provide a search result for a query that is synonymous with the actually input query. However, when a search result for a synonym query is provided, an unnecessary search result is provided because it is recommended as a query word in a synonym relationship even though the association with the actually entered query word is low.

Therefore, in this case, the user has to re-enter the search query in order to obtain a desired search result, which causes inconvenience to the user using the search service, which may lead to a decrease in satisfaction and reliability of the search service. .

An object of the present invention is to provide a search service providing method and system using synonym filtering that can filter synonyms using salmon information of synonyms.

Another object of the present invention is to provide a method and system for providing a search service using synonym filtering that can provide a search service using the filtered synonym.

Search service providing method using synonym filtering according to an aspect of the present invention for achieving the above object comprises the steps of extracting the salmon information of the query words belonging to the synonym group; Determining an association degree of query words belonging to the synonym group using the extracted salmon information; And filtering query terms belonging to the synonym group according to the determined association degree.

In an embodiment, the query words belonging to the synonym group may be query words in at least one of a heterogeneous form, a typo, invert, a tuning letter, and an abbreviation.

In this case, the extracting the salmon information may include extracting the salmon information from words input into a search box or extracting the salmon information from a web document.

The determining of the correlation may include determining the correlation using the salmon information matching ratio of the query words. According to an embodiment, the query word having the highest query count among query words belonging to the synonym group may be selected as the representative query word.

In an exemplary embodiment, the filtering of the query terms may include filtering a query word having a matching ratio of the representative query word and the salmon information among the query words belonging to the synonym group below a reference value.

In another embodiment, the filtering of the query words may include filtering the query words having a matching ratio of a representative query word among the query words belonging to the synonym group and specific salmon information among the salmon information to be lower than a reference value.

In this case, after the filtering step, the method may further include matching the filtered query terms with synonyms and storing the synonyms in a synonym database.

The search service providing method using the synonym filtering may include: receiving a search query word from a user; Determining whether a synonym of the received search query word exists; And providing a search result for the synonym in the search result for the received search query if the synonym of the received search query word exists.

Search service providing system using synonym filtering according to another aspect of the present invention for achieving the above object is a salmon information extraction unit for extracting the salmon information of the query words belonging to the synonym group; And a filtering unit for determining an association degree of the query words belonging to the synonym group by using the extracted salmon information and filtering the query words belonging to the synonym group according to the determined association degree.

As described above, according to the present invention, synonyms can be filtered more accurately and easily by using salmon information of query words determined as synonyms.

In addition, according to the present invention, when a search query is input from the user, the user's convenience is provided by including the search result for the filtered query among the queries determined as synonyms of the search query input in the search result for the input search query. As well as increase, this has the effect of improving the satisfaction and reliability of the search service.

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

1 is a block diagram illustrating a synonym filtering unit according to an embodiment of the present invention. The synonym filtering unit illustrated in FIG. 1 filters query terms belonging to a synonym group using salmon information of a corresponding query term. The synonym filtering unit is directly included in a search service providing system using synonym filtering, which will be described later, or a search service using synonym filtering. It may be configured independently of the providing system.

As shown, the synonym filtering unit 100 includes a salmon information extraction unit 120 and the filtering unit 140.

The salmon information extracting unit 120 extracts salmon information of query words belonging to a synonym group, and includes a first salmon information extracting unit 122 and a second salmon information extracting unit 124. Here, the salmon information basically means information about a word that is used with a certain distance in a sentence. This is because the query words belonging to the synonym group will form a plurality of common salmon information sets among the salmon information.

In an embodiment, the query words belonging to the synonym group may be query words in at least one of the following forms: typos, typos, inverts, tuning words, and abbreviations. This is because, among the query words input from the user by the synonym determination unit (not shown), the query words having at least one relation among this form, typo, invert, tuning fork, and abbreviation form a synonym group. Here, the synonym determination unit (not shown) may be directly included in the search service providing system using synonym filtering, which will be described later, like the synonym filtering unit 100, or may be configured independently from the search service providing system using synonym filtering.

here. Queries in a heterogeneous relationship mean a query word in which a vocabulary indicating the same object is expressed differently according to a user. In addition, the query words in the typo-related relationship are "cyworld," such as "cyworld," and in the case of "cyworld," which contains a phoneme that cannot form a syllable, or in a phoneme unit, such as "dang." In case of unscrup- ing, it means the query forms that the user inputs incorrectly, such as the case of the original notation "next" and the phoneme unit. In addition, the query words in the inverted relation mean, for example, query terms whose positions differ from each other in morphological units such as "Gangnam Kyobo" and "Kyobo Gangnam". In addition, query words in a tuning-fork relationship mean query words in a relationship between a Korean notation close to the original pronunciation of a foreign language such as "cyworld" and "Cyworld". Finally, the query terms in the abbreviation relationship refer to query terms in the relationship between the original word and the abbreviated word such as "Super Junior" and "Shuju".

An example of salmon information extracted by the salmon information extracting unit 120 is shown in FIG. 2. As shown in (a) of FIG. 2, the salmon information extracted for each of the query words "part-time", "part-time", and "arbat" can be known. In addition, as shown in (b) of FIG. 2, the salmon information extracted for each query "TV", "TV", "TV" can be known.

Referring back to FIG. 1, the first salmon information extracting unit 122 extracts salmon information from words input to a search box. In other words, it extracts information of words that are used at a certain distance within the query words that are words input by the user in the search box. For example, when a query such as "part-time job resume hourly site" is input by the user, the salmon information of the query "part-time job" may be "resume", "hourly hour", "site" and the like.

The second salmon information extracting unit 124 extracts salmon information from the web document. That is, it extracts information of words used together at a certain distance from a general web document. For example, if the web document contains the sentence "you can select a channel to watch and watch television schedules", the salmon information of the query "television" can be "organization table", "watch", "channel", and the like. .

The filtering unit 140 determines the association degree of the query words belonging to the synonym group by using the salmon information extracted by the salmon information extraction unit 120, and filters the query words belonging to the synonym group according to the determined association degree. The first filtering part 142 and the second filtering part 144 are included.

In an embodiment, the filtering unit 140 may determine the degree of association of the query words belonging to the synonym group by using the matching ratio of the salmon information corresponding to the query words belonging to the synonym group.

An example in which the degree of association of the query words belonging to the synonym group is determined by the filtering unit 140 using the match rate of the salmon information is illustrated in FIG. 3. As shown in (a) of FIG. 3, in the case of the query words "part-time" and "part-time", there is a large number of matching salmon information such as "heaven" and "site", whereas the query word "arbat" In case of, there is almost no salmon information that matches the query words "part-time" and "part-time". In addition, as shown in (b) of FIG. 3, in the case of the query words "television" and "television", a plurality of matching salmon information such as "organization table", "screen", "program", and "viewing" exist and are related. While the degree is high, the query term "television" has little salmon information that matches the query terms "television" and "television."

In an embodiment, the filtering unit 140 may select a query having the highest query count among the query words belonging to the synonym group as the representative query word. In other words, if the query word belonging to the query group determined as a synonym is "part-time", "part-time", "arbat", the "part-time" can be selected as the representative query if the number of query inputs of "part-time" is the highest. .

The first filtering unit 142 filters query queries having a matching ratio between the representative query word and the salmon information among the query words belonging to the synonym group below the reference value. In other words, the first filtering unit calculates a coincidence rate of the entire query information of the representative query word and the other query words, and excludes the query word having a match rate of the representative query word and the salmon information below the reference value from the synonym group.

The second filtering unit 144 filters the query words having a matching ratio of the representative query word among the query words belonging to the synonym group and the specific salmon information among the salmon information less than the reference value. In other words, the second filtering unit calculates a matching rate of specific salmon information that is important for determining correlation among the salmon query information of the representative query word and other query words, and excludes the query word whose matching rate between the representative query word and the specific salmon information is lower than the reference value from the synonym group.

As described above, the synonym filtering unit according to an embodiment of the present invention serves to refine the query words belonging to the synonym group in order to provide more accurate search results.

Hereinafter, a search service providing system using synonym filtering according to an embodiment of the present invention for providing a search service using the synonyms filtered by the synonym filtering unit will be described with reference to FIG. 4.

As illustrated, the search service providing system 600 using the synonym filtering includes the synonym filtering unit 100, the synonym database 200, the search query receiving unit 300, the synonym determination unit 400, and the search result providing unit 500. ).

The synonym filtering unit 100 filters the salmon information of the query words belonging to the synonym group to provide more accurate search results, and the synonym database 200 is performed by the synonym filtering unit 100 among the query words belonging to the synonym group. Filtered queries are stored.

The search query receiver 300 receives a search query from a user, and the synonym determination unit 400 determines whether a synonym of the search query received through the search query receiver 300 exists in the synonym database 200. In other words, the synonym determination unit determines whether a query entered in the relation of the search query input by the user and any one of a morphology, a typo, a doger, a tuning fork, and an abbreviation is stored in the synonym database 400.

The search result providing unit 500 includes a search result for the synonym of the received search query in the search result for the received search query when the synonym of the received search query exists in the synonym database 200. . Therefore, the search result providing unit 500 may provide not only search results for the search query input by the user but also search results in synonyms with the input search query, thereby providing more accurate search results.

Although the synonym filtering unit 100 is described as being included in the search service providing system 600 in the present embodiment, as mentioned above, the synonym filtering unit 100 is not included in the search service providing system 600. Maybe not. In this case, the search service providing system 600 may provide a search service by using the synonym database 400 in which query words filtered among the query words belonging to the synonym group by the synonym filtering unit 100 are stored.

Hereinafter, a synonym filtering method according to an embodiment of the present invention will be described with reference to FIG. 5.

First, a synonym group is created (S710). In other words, a group is created by determining the query words which are in one of the form, typo, invert, tuning fork, and abbreviation among the query words as synonyms. Accordingly, the query words belonging to the synonym group may be query words in at least one of the following forms: typos, typos, inverts, tuning words, and abbreviations.

Next, salmon information of query words belonging to the synonym group is extracted (S720). According to an embodiment, salmon information of query words belonging to a synonym group may be extracted from words input into a search box, or salmon information of query words belonging to a synonym group may be extracted from a web document.

Here, the salmon information basically means information about a word that is used with a certain distance in a sentence. In other words, since the query words belonging to the synonym group will form a plurality of common salmon information sets among the salmon information, the salmon information of the query words belonging to the synonym group is extracted.

As a result of the extraction, the degree of association between the query words belonging to the synonym group is determined using the extracted salmon information (S730). In one embodiment, the method of determining the degree of association may determine the degree of association using the salmon information matching rate of the query words belonging to the synonym group. In other words, when there is a lot of common salmon information among the salmon information of the query words belonging to the synonym group, it can be said that the correlation is high, but when there is almost no common salmon information, it can be said that the correlation is low.

In addition, among the queries belonging to the synonym group, the query having the highest query count number may be selected as the representative query. For example, when the query word belonging to the query group determined as a synonym is "part-time", "part-time", "arbat", the "part-time" may be selected as the representative query word if the number of query inputs of the "part-time" is the highest.

As a result of the determination, the query words belonging to the query group determined as synonyms according to the determined degree of association are filtered (S740). According to an embodiment, the method of filtering query terms may filter query terms having a representative query term and a salmon information matching rate below a reference value among query terms belonging to a synonym group. That is, the matching rate of the entire query information of the representative query and the other query terms is calculated, and the query terms whose match rate of the representative query information and the salmon information are less than the reference value are filtered out from the synonym group.

In another embodiment, the method of filtering the query words may filter query words having a matching ratio of the representative query word among the query words belonging to the synonym group and the specific salmon information among the salmon information less than the reference value. In other words, the coincidence rate of specific salmon information, which is important for determining correlation among the representative query word and other query words, is calculated, and the query word whose match rate of the representative query word and the specific salmon information is lower than the reference value is filtered in the synonym group.

Finally, a synonym database is constructed by storing the filtered query words among the query words belonging to the query group determined as the synonyms in the synonym database (S750). That is, the synonyms are determined among the query words input by the user, and the query words filtered by the salmon information of the query words belonging to the synonym group are stored in the synonym database.

Hereinafter, a method of providing a search service using a synonym filtered using the synonym filtering method described above will be described with reference to FIG. 6.

First, a search query is received from a user (S810), and it is determined whether a synonym of the received search query exists in the synonym database (S820). In other words, it is determined whether the query word entered by the user and the query word in one of the relational form, typos, vertebrae, tuning fork, and abbreviation relation are stored in the synonym database.

Next, when there is a synonym of the received search query, in providing a search result for the received search query, the search result for the received search query is included in the search result for the received search query. (S830). Therefore, not only the search result for the search query input by the user but also the search result for the synonym of the input search query can be provided together, thereby providing a more accurate search result.

Meanwhile, the search service providing method using the synonym filtering described above may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer-readable recording medium. In this case, the computer-readable recording medium may include program instructions, data files, data structures, and the like, alone or in combination. Meanwhile, the program instructions recorded on the recording medium may be those specially designed and configured for the present invention, or may be known and available to those skilled in computer software.

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. Magnetic-Optical Media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. The recording medium may be a transmission medium such as an optical or metal wire, a waveguide, or the like including a carrier wave for transmitting a signal specifying a program command, a data structure, or the like.

In addition, 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. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

On the other hand, those skilled in the art will appreciate that the present invention can be implemented in other specific forms without changing the technical spirit or essential features.

Therefore, it is to be understood that the embodiments described above are exemplary in all respects and not restrictive. The scope of the present invention is shown by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. do.

1 is a block diagram illustrating a synonym filtering unit according to an embodiment of the present invention.

2 is a diagram illustrating an example of extracted salmon information.

3 is a diagram illustrating an example in which an association degree of query words belonging to a synonym group is determined.

4 is a schematic block diagram of a search service providing system using synonym filtering according to an embodiment of the present invention including the synonym filtering unit illustrated in FIG. 1.

5 is a flowchart showing a synonym filtering method according to an embodiment of the present invention.

6 is a flowchart illustrating a search service providing method using synonym filtering according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

100: synonym filtering unit 120: salmon information extraction unit

122: first salmon information extracting unit 124: second salmon information extracting unit

140: filtering unit 142: first filtering unit

144: second filtering unit

Claims (21)

Extracting salmon information of query words belonging to the synonym group; Determining an association degree of query words belonging to the synonym group using the extracted salmon information; And And searching for query words belonging to the synonym group according to the determined degree of association. The method of claim 1, wherein the query words belonging to the synonym group Method for providing a search service using synonym filtering, characterized in that the query words in at least one of the form, typo, invert, tuning fork, abbreviation. The method of claim 1, wherein the extracting the salmon information Search service providing method using synonym filtering, characterized in that for extracting the salmon information from the words entered in the search box. The method of claim 1, wherein the extracting the salmon information Search service providing method using synonym filtering, characterized in that for extracting the salmon information from a web document. The method of claim 1, wherein the determining of the degree of association Method for providing a search service using synonym filtering, characterized in that the association degree of the query words belonging to the synonym group using the salmon information matching rate of the query words. The method of claim 1, Method for providing a search service using synonym filtering, characterized in that for selecting the highest query count query query among the query words belonging to the synonym group as the representative query word. The method of claim 1, wherein the filtering of the query terms A method of providing a search service using synonym filtering, the method comprising: filtering a representative query among the query words belonging to the synonym group and query words having a matching ratio of salmon information less than a reference value. The method of claim 1, wherein the filtering of the query terms Method for providing a search service using a synonym filtering, characterized in that the query between the representative query among the synonyms group and the specific information of the salmon information of the salmon information is less than the reference value. The method of claim 1, wherein after the filtering step, And matching the filtered query terms with synonyms and storing the filtered query terms in a synonym database. The method of claim 1, Receiving a search query from a user; Determining whether a synonym of the received search query word exists; And If there is a synonym of the received search query, the search service providing method using synonym filtering further comprising the step of providing a search result for the synonym in the search results for the received search query. . A recording medium having recorded thereon a program for carrying out the method according to any one of claims 1 to 10. A salmon information extraction unit for extracting salmon information of query words belonging to a synonym group; And Searching for synonym filtering using the extracted salmon information includes a filtering unit for determining the degree of association of the query words belonging to the synonym group, and filtering the query words belonging to the synonym group according to the determined association degree Service delivery system. The method of claim 12, wherein the query words belonging to the synonym group Search service providing system using synonym filtering, characterized in that the query words in at least one of the form, typo, invert, tuning fork, abbreviation. The method of claim 12, wherein the salmon information extraction unit And a first salmon information extracting unit for extracting the salmon information from words input in a search box. The method of claim 12, wherein the salmon information extracting unit And a second salmon information extraction unit for extracting the salmon information from a web document. The method of claim 12, wherein the filtering unit The search service providing system using synonym filtering, characterized in that the association degree is determined using the matching rate of the salmon information corresponding to the query words. The method of claim 12, wherein the filtering unit Search service providing system using synonym filtering, characterized in that for selecting a query with the highest query count (Query Count) among the query words belonging to the synonym group. The method of claim 12, wherein the filtering unit And a first filtering unit configured to filter query words having a matching ratio of the representative query word and the salmon information among the query words belonging to the synonym group equal to or less than a reference value. The method of claim 12, wherein the filtering unit And a second filtering unit configured to filter query words having a matching ratio of a representative query word among the query words belonging to the synonym group and specific salmon information among the salmon information equal to or less than a reference value. The method of claim 12, And a synonym database in which the filtered query terms are matched and stored as synonyms. The method of claim 12, A query receiving unit which receives a search query from a user; A synonym determination unit that determines whether a synonym of the received search query word exists; And If there is a synonym of the received search query, the search service using a synonym filtering further comprises a search result providing unit for providing a search result for the synonym in the search results for the received search query Provide system.

Images