KR100884889B1 - Method and system for adding automatic indexing word to search database - Google Patents

Abstract

The present invention relates to a method and system for automatic index addition of a search database. According to an aspect of the present invention, there is provided a method of automatically adding an index word to a search database, the method comprising: receiving a first search query from a user, providing at least one first result document indexed to the first search query from a search database, Receiving a second search query following the input of the first search query from the user, providing at least one second result document indexed into the second search query from the search database, wherein the at least Receiving a selection request for at least one result document of one second result document and adding the first search query to the search database as an index of the selected at least one second result document.

Search database, search query, index word, user log tracking

Description

METHOOD AND SYSTEM FOR ADDING AUTOMATIC INDEXING WORD TO SEARCH DATABASE}

1 is a diagram for explaining a result document provided to a user when a structure of a search database and a search query are input in the present invention.

2 is a diagram for a process of adding an automatic index word of a search database based on user log tracking according to one embodiment of the present invention.

FIG. 3 (a) illustrates a case in which a first search query is added as an index of a second result document selected by a user when a correlation exists between the first search query and the second search query according to an embodiment of the present invention. The figure which shows.

FIG. 3 (b) is a diagram illustrating a case in which a first search query is temporarily stored as a potential index word and added to the index word later when the inspection criteria is met.

FIG. 3 (c) automatically adds a first search query directly to an index word of a second result document when there is no click for a certain period of time for the first search query added to the index word. The figure which shows the case where it disappears.

4 is a diagram illustrating a process of adding a first search query to an index word of a second result document selected by a user according to one embodiment of the present invention.

FIG. 5 is a diagram illustrating a system for automatically adding an index word to a search database according to an embodiment of the present invention.

Description of the main parts of the drawing

502: search database

503: search query input unit

504: result document provider

505: selection request receiving unit

506: index word addition unit

The present invention relates to a method and system for automatically adding an index to a search database. More specifically, when a user inputs a search query and searches the user, the search query input by the user in a result document stored in the search database using the user's log tracking is searched. And a method for adding a to a new index word.

When a user inputs a search query in a search box to search, an appropriate search result may be exposed when a search query having a search result in which a result document for the search query is entered in a search database is used as an index word. However, in the case of a so-called internally built database that is indexed by an editor managing the system, it is unexpectedly overlooked to process an appropriate index word in a result document or responds appropriately to a user's accurate and fast search request due to an error. There was a problem that can not.

For example, if a user who wants to search for "badminton racket" enters "racquet" as a search query, the search result is displayed as a result document consisting only of "tennis racket", "squash racket" and "table tennis racket". Again, the trouble was to search for a combination of badminton and racket together. In particular, in the above example, when "racket" and "badminton racket" are related to the search query to some extent, as in the upper and lower relations, if only one of the search queries is treated as an index of the result document, the search result is narrowed. As a result, it was difficult to expose more detailed results, which resulted in a problem of satisfying the user's desired search result.

Even if a large number of result documents are stored in the search database, users may be forced to repeatedly search if the indexing process is not done properly or is missing. Problems arise that require time and effort.

The present invention proposes a method for adding an automatic index to a search database using a user log tracking and a new technique for a system for performing the method to solve the above problems of the prior art.

According to the present invention, when a user selects a result document exposed for a search query input by a user, the search query is added to the result document as an index word, and when a search of another user is made later, a more accurate and wider search result can be obtained. It aims to provide.

Another object of the present invention is to improve the user's satisfaction with the search as the classification is subdivided by adding index terms in the chain of different result documents.

In order to achieve the above object and to solve the problems of the prior art, the method for adding an automatic index of the search database according to an embodiment of the present invention, receiving a first search query (query) from the user, the first search query from the search database; Providing at least one first result document indexed by a search query, receiving a second search query following the input of the first search query from the user, and indexing the second search query from the search database Providing at least one second result document, receiving a selection request for at least one result document of the at least one second result document from the user, and transmitting the first search query to the search database. And adding to the index word of the at least one selected second result document.

According to an aspect of the present invention, the step of adding the first search query to the search database as an index of the selected at least one second result document has a correlation between the first search query and the second search query. The method may include determining whether the information is true and adding the first search query to the search database as an index of the selected at least one second result document when the correlation is established.

An automatic index word adding system of a search database according to an embodiment of the present invention includes a search database for storing and maintaining an index word and a result document, a search query input unit for receiving and storing a first search query and a second search query from a user, and A result document providing unit for providing at least one first result document indexed with the first search query from a search database and at least one second result document indexed with the second search query, the at least one first document from the user 1 is a selection request receiver for receiving a selection request for at least one of a result document and a second result document, and an index word for adding the first search query as an index of the at least one selected second result document to the search database. It may include wealth.

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

1 is a diagram for explaining a result document provided to a user when a structure of a search database and a search query are input in the present invention.

As shown in FIG. 1, the search database 101 is composed of a result document database 102, and a result document 103 which is a unit database is stored therein. In other words, the search database 101 may be referred to as a large database formed by gathering respective unit databases.

Each result document 103 has an index word (Tag), which is a word that can be exposed to a search result as information, and when a user inputs a search query, the search robot collects the result document inside the search database 101. Searching 103 may provide the user with appropriate search results. For example, when the user enters query2 105 as a search query in the search box of the search interface 104, the query2 105 is recognized as an index term 2 and the result of the search interface 104 is stored in the search database 101. D5, D6, D7, and D8, which are result documents 106 having index word 2 as information in the result document 103, are exposed. If the exposed result document 106 is the desired result, there is no problem when clicking to select it, but if the desired result is not exposed, there is a problem of re-searching by inputting another search query, and the present invention solves the problem. It is to improve the user's search satisfaction by adding a search query to the index.

2 is a diagram for a process of adding an automatic index word of a search database based on user log tracking according to one embodiment of the present invention.

According to the present invention, it can be seen that the user inputs a second search query consecutively within a short time after the user inputs the first search query to some extent between the two search queries.

In step S201, a first search query is first input from a user, and in step S202, a first result document indexed as a first search query from the search database 101 is provided to the user. In this case, the number of first result documents may vary depending on the result document database 102 indexed by the first search query of the search database 101 and may be one or plural.

In step S203, the second search query is continuously input to the first search query from the user, and in step S204, the second result document indexed as the second search query is searched from the search database 101 to the user. to provide. Here, in one embodiment, the user may select the first result document and subsequently enter the second search query to search. However, in view of the object of the present invention, the user may select the first result document and index the first result document. It is more likely that a second search query is input continuously without a click. In this case, the number of second result documents may vary depending on the result document database 102 indexed by the second search query of the search database 101.

In step S205, a selection request for at least one result document from the second result document is received from the user. In other words, the user selects a desired result document from the provided result documents. In this case, one or more result documents may be selected.

In operation S206, the first search query may be added to the search database as an index of the selected second result document. Even if a user selects a second result document by inputting a second search query successively without selecting a result document for the first search query, it is not always considered that there is some relation between the two search queries. The question arises whether it is possible to add a search query directly to the index word of the second result document. Therefore, when determining whether to add, it is necessary to set and adjust appropriate criteria. Reference settings and methods for adding the first search query to the index word of the second result document selected by the user will be described in detail with reference to FIGS. 3A, 3B, and 3C.

FIG. 3 (a) illustrates a case in which a first search query is added as an index of a second result document selected by a user when a correlation exists between the first search query and the second search query according to an embodiment of the present invention. The figure which shows. Referring to step S206 in more detail, it is determined whether a correlation is established between the first search query and the second search query (S301), and if the correlation is established between the two search queries, In operation S302, the first search query may be added to the index word of the selected second result document.

In one embodiment, there are four ways to determine whether there is a correlation. First, it can be said that a correlation is established when the first search query and the second search query satisfy an associated query condition. The association query condition generates the search query statistical information by measuring the frequency according to various features among the search query, the search query, the search query classification, or the result document classification associated with the result document set, and uses the vector, random variable, and combined probability distribution. By formulating and interpreting search query statistics information, the relationship between the search queries can be expressed as a numerical search query association, and the degree of association between the first search query and the second search query is constant based on the search query association. If the value is higher than the reference value, it may be determined that the correlation query condition is satisfied and the correlation is established between the two search queries.

Second, correlation may be established when the first search query includes the second search query or when the second search query includes the first search query. It refers to the case where there is an inclusive relationship between both search queries. For example, when the first search query is "Latin dance" and the second search query is "Latin dance group," the first search query includes the second search query. The correlation between the two can be judged. In addition, when the first search query is a "camera lens" and the second search query is a "camera peripheral", since the second search query includes the first search query, it may be determined that the correlation therebetween is also established.

Third, it can be said that the correlation holds when the first search query and the second search query include the same word. For example, when the first search query is "ballard music" and the second search query is "ballard song", it is determined that the correlation is established because the first search query and the second search query include the same "ballard". can do.

Fourth, it can be said that a correlation is established when a search pattern from the first search query to the second search query occurs at a frequency equal to or greater than a predetermined reference value. The reference value is a numerical value by dividing the pattern for searching by inputting the first search query followed by the second search query by time, daily, monthly, and year. In this case, there should be a meaningful search pattern, which assumes that there is some relation between the two search queries.

 3 (b) is a diagram illustrating a case in which a first search query is not immediately added as an index word but temporarily stored as a potential index word and added as an index word when satisfying a test criterion later. to be. More specifically, in step S206, storing the first search query as a potential index word of the selected second result document in a search database, and storing a potential index word satisfying a predetermined inspection criterion in the selected second result document. It can be seen as a step (S304) of adding to the index word. In other words, instead of automatically adding the first search query as an index, it stores the potential search term in the potential index term field for later addition, and periodically or irregularly stores the potential index term as a checker within the database that meets certain criteria. You can add At this time, the frequency of the potential index word actually used as the index word of the second search document can be determined by a certain inspection criterion.

FIG. 3 (c) automatically adds a first search query directly to an index word of a second result document when there is no click for a certain period of time for the first search query added to the index word. The figure which shows the case where it disappears. If the first search query is not selected for a predetermined period of time in step S206, the step S305 may be added to delete the index word from the search database. It is a post-validation method that can be added to the index word first, and then automatically deleted if the number of times the user enters the first search query to search the second result document for a certain period is less than the criteria set in the database. .

4 is a diagram illustrating a process of adding a first search query to an index word of a second result document selected by a user according to one embodiment of the present invention.

First, when a user enters a first search query query1 in a search box, a search result 401 for the first result document is displayed. In the search database 101 described in FIG. 1, since the result documents classified by index word 1 are stored as D1, D2, D3, and D4, the result documents processed by index word 1 corresponding to query1 are D1, D2, D3, and D4. Appears as a search result 401 for the first result document. At this time, it is assumed that the user does not respond to clicking on the first result documents.

When a user continuously inputs a second search query query2 into a search box, a search result 402 for the second result document is displayed. In the search database 101, the result documents D5, D6, D7, and D8 processed by the index term 2 are displayed as the search result 402 for the second result document. At this time, it is assumed that the user clicks on D6 and D8 to select.

Then, it can be seen that the first search query is added to the index word for D6 and D8 selected by the user in the result document database 403 processed by the index word 2 of the search database 101. In this case, a specific method of adding the first search query to the index word of the second result document selected by the user may be classified according to the detailed conditions described with reference to FIGS. 3A, 3B, and 3C.

FIG. 5 is a diagram illustrating a system for automatically adding an index word to a search database according to an embodiment of the present invention.

The automatic index word adding system 501 of the search database includes a search database 502, a search query input unit 503, a result document providing unit 504, a selection request receiving unit 505, an index word adding unit 506, and a potential index fish unit (506). 507, and an inspector 508.

The search database 502 may function to store and maintain a result document in which index terms are processed. Specifically, referring to FIG. 1, the search database 502 includes a result document database 102. The result document database 102 stores a result document 103 that is a unit database. In other words, the search database 502 may be referred to as a large database formed by gathering respective unit databases. In addition, rather than adding a search query immediately as an index word of the result document, a potential index part 507, which is a space for temporarily storing the potential index word, and a check part 508 for determining whether the potential index word satisfies a certain inspection criteria may be included. have.

The search query input unit 503 may receive and store a search query by providing a search interface to the user 509 in the automatic index word adding system 501 of the search database. Based on the user log tracking, the search query may be continuously input from the user 509. The result document providing unit 504 may provide a result document in which an index word corresponding to a search query received from a user has been processed, from the search database 502.

In other words, it serves to provide the user with at least one result document indexed by the search query from the search database 502. In this case, based on the user log tracking, all the result documents for each successive search query may be provided. The provided result document may be one or plural depending on the state stored in the search database.

The selection request receiving unit 505 serves to receive the selection request for the document selected by the user 509 among the provided result documents. Based on the user log tracking, the user can receive the selection request of the user 509 for the result documents for the first search query and the second search query, and if the classification or the related search query are used, the search query entered by the user can be received. A selection request of the user 509 may be received for the result document. In general, the user's selection request may be that the user 509 clicks on the result document, and there may be one or a plurality of selection requests among the provided result documents.

The index word adding unit 506 adds a search query to the search database 502 as an index word of the result document selected by the user.

When adding an index word based on user log tracking, the first search query may be added as an index word of at least one selected second result document. In this case, the index word adding unit 506 may be divided into three types, and the specific method may be referred to in FIGS. 3A, 3B, and 3C. Here, the correlation determination unit in the system may be in charge of determining whether a correlation is established between the first search query and the second search query.

The potential index unit 507 functions to temporarily store the search query as a potential index rather than storing the search query as an index of the result document directly in the search database 502. In other words, the search query input by the user may be a space temporarily added to the index word in order to find out whether a certain inspection criterion is satisfied before adding it to the search database 502 as the index word.

The inspection unit 508 determines whether the potential index word stored in the potential index word unit 507 satisfies a predetermined inspection criterion on the system regularly or irregularly. It adds index words.

The automatic index word adding method of the search database according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on the media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. 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. 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.

As described above, although the present invention has been described with reference to limited embodiments and drawings, the present invention is not limited to the above embodiments, and those skilled in the art to which the present invention pertains various modifications and variations from such descriptions. This is possible.

Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the claims below but also by the equivalents of the claims.

According to the present invention, it is possible to provide a method and system for automatically indexing a search database using user log tracking.

According to the present invention, when there is a selection of the result document exposed for the search query input by the user, the search query is added to the result document as an index word, and when there is another user's search later, a more accurate and wider search result Can be provided.

According to the present invention, a search query is sequentially added to a search database as an index word of a result document, and as the classification is further segmented, the user's satisfaction with the search is improved.

Claims (19)

Receiving a first search query from a user; Providing at least one first result document indexed by the first search query from a search database; Receiving a second search query following the input of the first search query from the user; Providing at least one second result document indexed by the second search query from the search database; Receiving a selection request for at least one result document of the at least one second result document from the user; And Adding the first search query to the search database as an index of the selected at least one second result document. How to add an automatic index of the search database that includes. The method of claim 1, The step of receiving a second search query following the input of the first search query from the user, Receiving a second search query from the user without a selection request of at least one first result document indexed by the first search query from the user How to add an automatic index of the search database that includes. The method of claim 1, The adding of the first search query to the search database as an index of the selected at least one second result document may include: When the first search query includes the second search query or when the second search query includes the first search query, a correlation is established between the first search query and the second search query. Judging by; And Adding the first search query to the search database as an index of the selected at least one second result document when correlation is established. How to add an automatic index of the search database that includes. The method of claim 3, Determining that a correlation is established when the first search query and the second search query satisfy an associated query condition of the search database. More, The association query condition, Automated search database characterized in that it is a condition for determining whether the search query association between the first search query and the second search query is greater than a predetermined reference value according to the statistical information on the search query frequency How to add index terms. delete The method of claim 3, Determining that a correlation is established when the first search query and the second search query include the same word. How to add an automatic index of the search database that includes more. The method of claim 3, Determining that a correlation is established when a search pattern from the first search query to the second search query occurs at a frequency equal to or greater than a predetermined reference value. How to add an automatic index of the search database that includes more. The method of claim 1, The adding of the first search query to the search database as an index of the selected at least one second result document may include: Storing the first search query in the search database as a potential index of the selected at least one second result document; And Adding the potential index word that satisfies a certain inspection criterion to the index word of the selected at least one second result document. How to add an automatic index of the search database that includes. The method of claim 1, Deleting the index word from the search database when the first search query is not selected for a predetermined period of time. How to add an automatic index of the search database that includes more. A computer-readable recording medium having recorded thereon a program for executing the method of any one of claims 1 to 4 or 6 to 9. A search query input unit which provides a search interface to receive and store a first search query and a second search query from a user; A result document providing at least one first result document indexed with the first search query and at least one second result document indexed with the second search query from a search database that stores and maintains an index document processed result document Providing unit; A selection request receiving unit which receives a selection request for at least one of the at least one first result document and the second result document from the user; And An index word adding unit that adds the first search query to the search database as an index of the selected at least one second result document. Automatic indexing system of the search database that includes. The method of claim 11, The search query input unit, And there is no request for selection of at least one first result document indexed by the first search query from the user, and a second search query is input from the user successively. The method of claim 11, When the first search query includes the second search query or when the second search query includes the first search query, a correlation between the first search query and the second search query is established. Correlation judgment unit to judge More, The index word adding unit, If the first search query and the second search query have a correlation, the first search query is added to the search database as an index of the at least one selected second result document. Index word addition system. The method of claim 13, The correlation determination unit, If the first search query and the second search query satisfies the related query conditions of the search database, it is determined that a correlation is established, The association query condition, Automated search database characterized in that it is a condition for determining whether the search query association between the first search query and the second search query is greater than a predetermined reference value according to the statistical information on the search query frequency Index word addition system. delete The method of claim 13, The correlation determination unit, And if the first search query and the second search query contain the same word, determine that a correlation is established. The method of claim 13, The correlation determination unit, And a correlation is determined when a search pattern from the first search query to the second search query occurs at a frequency equal to or greater than a predetermined reference value. The method of claim 11, A potential indexer for storing and maintaining a search query as a potential index of the result document; And A inspecting unit that determines whether the potential index word satisfies a certain inspection criteria More, The index word adding unit, And the potential index word determined by the inspector to satisfy a predetermined inspection criterion is an index word of the selected at least one second result document. The method of claim 11, The index word adding unit, And the index word is deleted from the search database if the first search query is not selected for a predetermined period of time.

Images