Information suggesting method and system

Abstract

The invention relates to an information suggesting method and an information suggesting system. The method comprises the following steps of recording each search term selected by a user and corresponding result information, and setting the characteristic value of the result information; selecting first N pieces of result information from the result information which corresponds to each search term according to the characteristic value, wherein N is a natural number, and is greater than or equal to 2; selecting first M pieces of result information from the first N pieces of result information of each search term according to the characteristic value, wherein M is a natural number, and is less than N; partitioning each search term to a plurality of participle units, and establishing a corresponding relation between each participle unit and a plurality of search terms; and establishing a corresponding relation between each participle unit and the first M pieces of result information of a plurality of search terms respectively according to the corresponding relation between each participle unit and a plurality of search terms. Due to the adoption of the information suggesting method and the information suggesting system, the first M pieces of superior result information of each search term can be provided for the user in a retrieving process, the requirements of user are met directly, and the retrieving cost is lowered.

Description

technical field [0001] The invention relates to the field of information retrieval, in particular to an information suggestion method and system. Background technique [0002] Query suggestion, also known as Query Suggestion, figure 1 It is a schematic diagram of the interface of an existing query suggestion system (search engine), such as figure 1 As shown, when the user enters the search term "Huanzhu" in the input area of ​​the search engine, the search engine will automatically provide multiple query terms 11 in the form of a drop-down list, such as "Huanzhugege", "Huanzhugege this swallow". "Pianpianfei" and so on to improve the efficiency of users' retrieval of information. At present, the query suggestion system has become an important way to reduce the cost of users' information acquisition. For example, 30% of Baidu's search traffic comes from the suggestion system. Is the query suggestion system well done? , will directly affect the user experience. [0003] How...

AI Technical Summary

Problems solved by technology

[0002] Query suggestion, also known as Query Suggestion, figure 1 It is a schematic diagram of the interface of an existing query suggestion system (search engine), such as figure 1 As shown, when the user enters the search term "Huanzhu" in the input area of ​​the search engine, the search engine will automatically provide multiple query terms 11 in the form of a drop-down list, such as "Huanzhugege", "Huanzhugege this swallow". "Pianpianfei" and so on to improve the efficiency of users' retrieval of information. At present, the query suggestion system has become an important way to reduce the cost of users' information acquisition. For example, 30% of Baidu's search traffic comes from the suggestion system. Is the query suggestion system well done? , will directly affect the user experience

[0003] However, the traditional query suggestion system is mainly based on text display, which has limited information, is not friendly enough to interact with, and users need to search the page to find the page they want. Accordingly, some innovative approaches have emerged in the industry , such as combining entity databases (TV library information, movie library information) with a query suggestion system, figure 2 Suggest system interface diagrams for existing queries that incorporate entity repository information, such as figure 2 As shown, if the query words in the query suggestion system are matched in the entity library, the corresponding information will be displayed on the right side of the interface of the query suggestion system12, although this suggestion system combined with the entity library information can display More information makes the interaction richer, but its disadvantage is that only when the search term entered by the user exactly matches the standard query term preset by the query suggestion system, the corresponding information 12 in the entity database can be displayed. If If the user enters a search term similar to the standard query term, the corresponding information cannot be displayed12. In this case, there is not enough recall on the one hand, and on the other hand, it also brings confusion to the user. In addition, the entity database The corresponding information 12 in the database is usually one-to-one matching with the standard query words. Therefore, the current query suggestion system combined with the entity database information cannot display multiple pieces of corresponding information 12 to the user.

Images