WO2012106941A1

WO2012106941A1 - Method and device for full-text search

Info

Publication number: WO2012106941A1
Application number: PCT/CN2011/077788
Authority: WO
Inventors: 樊彪
Original assignee: 华为技术有限公司
Priority date: 2011-07-29
Filing date: 2011-07-29
Publication date: 2012-08-16
Also published as: CN102317943A; CN102317943B

Abstract

Disclosed are a method and device for full-text search, belonging to the field of information search. In the present invention, the contents of searchable documents are stored in an index library using a preset information model, and classification information is added, so that when the search engine outputs the search results, a broad category and a precise category to which the search results belong can be acquired according to the classification information so as to display the search results in a classified manner, and the user can quickly acquire the desired results by filtering according to the broad category and the precise category, and the user can carry out search more conveniently and quickly, reducing the search workload of the user.

Description

Method and device for full-text search

The present invention relates to the field of information search, and in particular, to a method and apparatus for full-text search. Background technique

With the rapid growth of information scale, how to say that efficient and accurate access to the information needed by users is becoming an urgent problem to be solved. At this stage, search engines are an effective means of solving the above problems. Most search engines use keyword matching methods to match the data in the repository with the keywords entered by the user to get the information the user needs.

Book

However, the inventors have found that the method for searching for information required by a user through a search engine in the prior art has the following drawbacks: The number of search results is huge, and the search results may involve many fields, and it is often difficult for users to find content of their own interest in search results. of. Summary of the invention

Embodiments of the present invention provide a method and apparatus for full-text search to solve the problems in information search in the prior art.

A method for full-text search, the method comprising:

Receiving a keyword input by the user, performing matching in the index library according to the keyword to obtain a search result; extracting, in the index library, classification information of the search result;

And obtaining, according to the classification information, a classification large item and a classification small item to which the search result belongs, and classifying the search result according to the classification large item and the classification small item to which the search result belongs;

All the searchable documents mapped by the preset information model are stored in the index library, and the preset information model includes the classification information of the all searchable documents and the all searchable documents. Title and body content. A device for full-text search, characterized in that the device comprises:

a search module, configured to receive a keyword input by the user, and perform matching according to the keyword in the index library to obtain a search result; a classification information obtaining module, configured to extract classification information of the search result in the index library; and a classification display module, configured to acquire, according to the classification information, a classification item and a classification item belonging to the search result, and according to the classification information The classified large item and the classified small item to which the search result belongs display the search result into categories;

All the searchable documents mapped by the preset information model are stored in the index library, and the preset information model includes the classification information of the all searchable documents and the all searchable documents. Title and body content. The beneficial effects of the technical solution provided by the embodiment of the present invention are as follows: In the embodiment of the present invention, the content of the searchable document is saved in the index database by using a preset information model, and the classification information is added, so that the search engine outputs the search result. According to the classification information, the classification item and the classification item belonging to the search result can be classified and displayed, so that the user can quickly obtain the desired result according to the classification big item and the classification item, and search for the user. It is more convenient and faster, reducing the workload of user search. DRAWINGS

1 is a schematic flowchart of a method for full-text search provided in Embodiment 1 of the present invention;

2 is a schematic flowchart of a method for full-text search provided in Embodiment 2 of the present invention;

3 is a schematic structural diagram of a classifier provided in Embodiment 2 of the present invention;

4 is a schematic diagram showing a display result of a search result after full-text search provided in Embodiment 2 of the present invention; FIG. 5 is a schematic diagram showing a sorted search result display situation provided in Embodiment 2 of the present invention;

6 is a schematic structural diagram of an apparatus for searching for a first full text according to Embodiment 3 of the present invention;

FIG. 7 is a schematic structural diagram of a second full-text search apparatus according to Embodiment 3 of the present invention; FIG.

8 is a schematic structural diagram of a third full-text search apparatus provided in Embodiment 3 of the present invention;

FIG. 9 is a schematic structural diagram of a fourth full-text search apparatus according to Embodiment 3 of the present invention. detailed description

The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

Example 1

As shown in FIG. 1 , an embodiment of the present invention provides a method for full-text search, where the method includes:

101. Receive a keyword input by a user, and perform matching according to the keyword in the index library to obtain a search result;

102. Extract classification information of the search result in the index library; 103. Acquire, according to the classification information, a classification large item and a classification small item to which the search result belongs, and classify and display the search result according to the classified large item and the classified small item to which the search result belongs.

All the searchable documents mapped by the preset information model are stored in the index library, and the preset information model includes the classification information of all searchable documents and the title and body content of all searchable documents. The full-text search method provided by the embodiment of the present invention saves the content of the searchable document in the index database by using a preset information model, and adds the classification information, so that the search engine can output the search result according to the classification information. The classification item and the classification item belonging to the search result are classified and displayed, so that the user can quickly obtain the desired result by filtering according to the classification item and the classification item, and the user is more convenient and fast to search. Reduce the workload of user search. Example 2

As shown in FIG. 2, the embodiment of the present invention provides a method for full-text search, and Embodiment 2 is a refinement based on Embodiment 1, to explain the method provided by the present invention.

A method for full-text search, specifically comprising:

201. Save the content of all searchable documents in an index library according to a preset information model;

The content of all searchable documents is stored in the index library according to a preset information model, specifically: mapping the content of the searchable document into a body content and a document of a set of metadata and searchable documents. The information consisting of the title. The metadata includes at least classification information of the searchable document.

It should be noted that the preset information model defines a custom information format, and the content of the searchable document can be mapped to a format defined by the preset information model, and the content of a searchable document is pre-predicted. The information model is saved as an example to illustrate:

<html>

<head>

</head> <title> Alarm Box Reporting a Critical Alarm When No Alarm Occurs</title><body>

Hel lo, World

</body>

</html>

In this embodiment, the content of the searchable document mapped by the information model is saved as an Html format as an example. In an actual application, the content of the searchable document mapped by the information model may also be used in other formats. Save it.

In the above information obtained by mapping the content of the searchable document through the preset information model, the content between "<h _ea d>" and 〃</head>" is metadata, "<meta name ⁼ DC. Type"content=〃reference〃>〃 and 〃 <meta name ⁼ DC. Audience. Job " content ⁼ trouble management 〃> 〃 is classification information, where "DC. Type" is the metadata name, indicating that the field is classification information "reference" is the value corresponding to the metadata "DC. Type", 〃 DC. Audience. Job〃 is also the metadata name, indicating that the field is classification information, 〃 trouble management "for 兀 data" DC. Audience. Job " Corresponding values, "Alarm Box Reporting a Critical Alarm When No Alarm Occurs 〃 and Hel Hel lo, World 〃 are the title and body content of the searchable document, respectively.

The index library may be a database, and is used to save information obtained by mapping the content of the searchable document through a preset information model.

Further, in the preset information model, a summary field, a keyword field, and the like may be further included, for example, information obtained after mapping by a preset information model, specifically:

Abstract field 〃 abstract〃为〃 This describes the symptoms.〃, keyword field 〃 keywords〃

"alarm box fault, alarm box reporting a critical alarm when no alarm occurs".

Further, in the preset information model, field expansion may also be performed, for example, adding a custom field to add classification information and a rule indicating the display of the classification item.

202. Establish a classifier according to a preset information model.

The corresponding relationship between the classified large item, the classified small item, and the classification information is stored in the classifier.

Specifically, the classifier is composed of one or more text files, and the structure thereof is composed of three parts: a classification item, a classification item, and metadata, and is illustrated by using FIG. 3 as an example, wherein the classification item includes a large item A. , B, N, classification items include small items Al, A2, AN, B B2, BN, Nl, and metadata includes VA1, VA2, VAN, VB VB2, VBN, and Li.

In the classifier structure diagram shown in Figure 3, corresponding to the preset information model, the classification large items are defined, and the classification small items are defined under each classification large item, and the classification information in the metadata is made. Corresponds to the classification item.

It should be noted that, in the classifier structure diagram shown in FIG. 3, each of the classification items and the preset information model Corresponding to the classification information in the metadata, the classification information is extracted from the metadata of the information model corresponding to each searchable document, and the classification item and the classification large item to which the searchable document belongs are determined according to the classifier.

Further, a searchable document may be defined with a plurality of classification information, and in this way, the searchable document is divided into a plurality of classified large items and classified small items. For example, if the classification information is read in the information model of the searchable document as VA2 and VB2, the searchable document belongs to the classification small item A2 of the classification large item A and the classification of the large classification item B when performing classification. Item B2.

Further, the classifier can perform custom extensions to add new classification items and/or classification items.

203. Create a document index and save it to an index library.

Wherein, the document index is established. Specifically, the search engine automatically extracts the classification information in the metadata of all the searchable documents saved according to the preset information model, and obtains the classification items and the classifications of all the searchable documents by the classifier. The item, and is saved as a document index in the index library.

Further, obtaining, by the classifier, the classification large item and the classification small item to which all the searchable documents belong, specifically: acquiring, according to the classification information in the metadata, the classification item belonging to all the searchable documents by the classifier, and according to the classifier The corresponding relationship between the classified large item and the classified small item saved in the category determines the classified large item to which all the searchable documents belong, and saves the classified large item and the classified small item to which all the searchable documents belong as the document index in the index library.

204. Receive a keyword input by a user, and perform matching according to the keyword in the index library to obtain a search result;

205. Obtain a classification item and a classification item belonging to the search result by using an index library;

The indexing library obtains the classified large item and the classified small item to which the search result belongs, and specifically, in the index library, the classified large item and the classified small item to which the search result saved in the document index belongs are read.

206. Display and display the search results according to the classified large items and the classified small items to which the search result belongs.

In this embodiment, the search results are classified and displayed according to the classification information, which specifically includes:

According to the classified large item and the classified small item to which the above search result belongs, the search result is displayed according to the classified large item and the classified small item classified by the search result.

In the search result display situation shown in FIG. 4, the searchable document included in the search result includes three categories of large-scale operation and maintenance processes, document types, product models, and operation and maintenance processes, document types, and product models. It is divided into several sub-items.

Further, the above method further includes:

The classification item selected by the user is received, and the search result included in the classification item is filtered according to the keyword, and the search result after the filtering search is performed in the searchable document included in the category item is displayed.

For example, as shown in the search result shown in FIG. 4, the search search may be performed based on the keyword among the 388 search results included in the classified item product A, and only the search results included in the classified item A may be displayed. Narrow down The range of search results is more convenient for users to get the closest search results.

Further, the method for the full-text search described above may further include:

Establishing a key segment weighter, defining different weights for different key fields in the key segment weighter, and calculating weighted results of all searchable documents mapped by the preset information model saved in the index library, The weighted result is saved to the index library as a weighted index.

The key field includes a content field and a classification field defined in a preset information model, a content field such as a title, a keyword field, and the like, a classification field such as a summary, a variety of custom fields, and the like, and a body content field of the searchable document may be As a standard field, the key segment weighter is specifically a text file, and the weight of the key segment can be weighted according to the key segment weighter. The greater the weighting result corresponding to the searchable document, the more relevant the searchable document is. High, the higher the priority of the search results output.

For example, the key segment weighters are described in Table 1:

Table 1

In Table 1, the title, keyword, and instruction command are all key fields defined in the preset information model. As the information model in step 201, body can be used as a standard field.

Correspondingly, after performing weighting calculation on all searchable documents mapped according to the preset information model, step 206 specifically includes: classifying all the search results according to the classification information, and displaying all the search results according to the relevance of all the search results. Sort display. As shown in FIG. 5, the ranking results of the search results after the key segment weighting process are displayed, and the top ranked is the most relevant search result, which can make the user get the accuracy of the search result and the user needs to search. Keep the document in front and avoid too many page-finding operations. The full-text search method provided by the embodiment of the present invention saves the content of the searchable document in the index database by using a preset information model, and adds the classification information, so that the search engine can output the search result according to the classification information. The classification item and the classification item belonging to the search result are classified and displayed, so that the user can quickly obtain the desired result by filtering according to the classification item and the classification item, and the user is more convenient and fast to search. Reduce the workload of user search. Example 3

An embodiment of the present invention provides a device for full-text search. As shown in FIG. 6, the device specifically includes:

The search module 301 is configured to receive a keyword input by the user, and perform matching according to the keyword in the index library to obtain a search result;

a classification information obtaining module 302, configured to extract classification information of the search result in the index library;

The classification display module 303 is configured to obtain, according to the classification information, the classification large item and the classification small item to which the search result belongs according to the classification information, and classify and display the search result according to the classification large item and the classification small item to which the search result belongs, and classify and display the search result;

Wherein, all the searchable documents mapped by the preset information model are stored in the index library, and the preset information model includes the classification information of all searchable documents and the body content of all searchable documents. Further, as shown in FIG. 7, the foregoing apparatus further includes:

The document indexing module 304 is configured to: before the search module 304 receives the keyword input by the user, establish a classifier according to the preset information model, and extract classification information in all searchable documents mapped by the preset information model, according to the classification And classification information to obtain the classification items and classification items belonging to all searchable documents, and save them as document indexes, and save the document indexes in the index library;

The classifier stores a correspondence relationship between the classification large item, the classification small item, and the classification information.

Correspondingly, the classification display module 303 is specifically configured to: obtain, according to the classification information of the search result, a classification item and a classification item belonging to the search result in the document index, and display the search according to the classification item and the classification item category to which the search result belongs result.

Further, as shown in FIG. 8, the foregoing apparatus further includes: The filtering search module 305 is configured to: after the classification display module 303 sorts and displays the search results according to the classification information, receive the classification item selected by the user, and perform filtering search according to the keyword in the searchable document included in the classification item, display filtering Search results after search.

Further, as shown in FIG. 9, the foregoing apparatus further includes:

The weighted index establishing module 306 is configured to establish a key segment weighter according to a preset information model before the search module 304 receives the keyword input by the user, and define different key segments for different key segments in the key segment weighter. Weight, calculate the weighted result of all searchable documents after the preset information model mapping, and store the weighted result as a weighted index in the index library.

Correspondingly, the classification display module 303 is specifically configured to classify and display the search results according to the classification information, and obtain the weighted results of the search results saved in the weighted index, and display the search results in descending order according to the weighting result. The apparatus for full-text search provided by the embodiment of the present invention saves the content of the searchable document in the index database by using a preset information model, and adds the classification information, so that the search engine can output the search result according to the classification information. The classification item and the classification item belonging to the search result are classified and displayed, so that the user can quickly obtain the desired result by filtering according to the classification item and the classification item, and the user is more convenient and fast to search. Reduce the workload of user search. It should be noted that: the apparatus for full-text search provided by the foregoing embodiment is only illustrated by the division of each functional module. In actual applications, the function allocation may be completed by different functional modules as needed, that is, full-text search. The internal structure of the device is divided into different functional modules to perform all or part of the functions described above. In addition, the device for the full-text search provided by the foregoing embodiment is the same as the method for the full-text search. The specific implementation process is described in detail in the method embodiment, and details are not described herein again.

All or part of the technical solutions provided by the above embodiments may be implemented by software programming, and the software program is stored in a readable storage medium such as a hard disk, an optical disk or a floppy disk in a computer.

The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., which are within the spirit and scope of the present invention, should be included in the protection of the present invention. Within the scope.

Claims

Claim

A method for full-text search, characterized in that the method comprises:

All the searchable documents mapped by the preset information model are stored in the index library, and the preset information model includes the classification information of the all searchable documents and the all searchable documents. Title and body content.

The method according to claim 1, wherein before the receiving the keyword input by the user, the method further includes:

Establishing a classifier according to the preset information model;

Extracting classification information in all searchable documents after the preset information model mapping;

Obtaining, according to the classifier and the classification information, a classification item and a classification item belonging to the all searchable documents, and saving the file index as a document index, and saving the document index in the index library;

The classifier stores a correspondence between the classification major item, the classification item, and the classification information.

The method according to claim 2, wherein the displaying the search results according to the classification information comprises:

Obtaining, according to the classification information of the search result, the classified large item and the classified small item to which the search result belongs, and displaying the search result according to the classified large item and the classified small item classified by the search result. .

The method according to claim 3, wherein, after the classifying the search results according to the classification information, the method further includes:

And receiving the classified item selected by the user, performing a filtering search according to the keyword in the searchable document included in the classified item, and displaying the search result after filtering the search.

5. The method according to any of claims 1-4, wherein the receiving user loses Before entering the keyword, the method further includes:

Determining a key segment weighter according to the preset information model, and defining different weights for different key fields in the key segment weighter;

Calculating a weighted result of all searchable documents after the preset information model mapping, and storing the weighted result as a weighted index in the index library.

The method according to claim 5, wherein the displaying the search results according to the classification information comprises:

The search results are classified and displayed according to the classification information, and the weighted results of the search results saved in the weighted index are obtained, and the search results are displayed in descending order according to the weighting result.

7. A device for full-text search, wherein the device comprises:

a search module, configured to receive a keyword input by the user, and perform a matching in the index library according to the keyword to obtain a search result;

a classification information obtaining module, configured to extract, in the index library, classification information of the search result;

a classification display module, configured to acquire, according to the classification information, a classification large item and a classification small item to which the search result belongs, and classify and display the search result according to the classification large item and the classification small item to which the search result belongs;

8. The device according to claim 7, wherein the device further comprises:

a document indexing module, configured to: before the searching module receives the keyword input by the user, establish a classifier according to the preset information model, and extract all the searchable documents that are mapped by the preset information model And the classification item is obtained according to the classifier and the classification information, and is saved as a document index, and the document index is saved in the index library;

The apparatus according to claim 8, wherein the classification display module is configured to: obtain, according to the classification information of the search result, a classification item and a classification to which the search result belongs according to the document index. Small item, and follow The classified large item and the classified small item classified by the search result display the search result.

The device according to claim 9, wherein the device further comprises:

a filtering search module, configured to: after the classification display module classifies and displays the search result according to the classification information, receive a classification item selected by the user, in a searchable document included in the classification item, according to The keyword performs a filtering search to display search results after filtering the search.

11. The device according to any one of claims 7-10, wherein the device further comprises:

a weighted index establishing module, configured to establish a key segment weighter according to the preset information model before the search module receives the keyword input by the user, and is different in the key segment weighter The key segment defines different weights, and the weighted result of all searchable documents after the preset information model mapping is calculated, and the weighted result is saved in the index database as a weighted index.

The device according to claim 11, wherein the classification display module is configured to: classify and display the search result according to the classification information, and acquire the search result saved in the weighted index The weighted result displays the search results in descending order according to the weighted result.