Method and device for recommending input association based on historical search of user

Abstract

The invention provides a method and a device for recommending input association based on historical search of a user. The method includes the steps: receiving search terms inputted by the user; acquiring a preset input association candidate set; acquiring hit historical search terms of the user if analyzing that parts or all of keywords of the current search terms hit historical search terms of the user in the input association candidate set; recommending the input association according to the acquired historical search terms of the user. The efficiency of the method for recommending the inputassociation based on historical search of the user is improved. The user can accurately think of proper search terms to express search intention without manually rewriting the search terms for searchagain, and accuracy of the personalized input association recommendation method can be improved. By the aid of an existing input association model, the search cost of the user can be reduced, the user can be helped to rapidly and conveniently find out search results, the deriving effect of a whole search box can be improved, and search experience is further improved.

Description

Technical field [0001] The present invention relates to the field of computer technology, in particular to a method and device for recommending input association based on user history search. Background technique [0002] With the rapid development of the Internet, people now increasingly like to use search engines to search for materials that are used for learning or interest. In the process of using the existing search engine, when the user enters manually, such as when the input is wrong, when the search result is clicked, any recommended word on the page is displayed. At this time, the user needs to manually rewrite the search term and search again. There are high-frequency queries that need to be searched repeatedly, such as official website addressing, audio and video, and Xu Ru’s complete pinyin bytes are required each time. Users cannot accurately think of appropriate search terms to express their search intentions, resulting in existing input associations The recommende...

AI Technical Summary

Problems solved by technology

In the process of using the existing search engine, when the user manually enters, such as an input error, after clicking the search result, any recommended words on the page will be displayed. At this time, the user needs to manually rewrite the search term to search again. If the user There are high-frequency queries that need to be searched repeatedly, such as official website addressing, audio-visual video, and Xu Ru’s complete pinyin bytes are required every time. Users cannot accurately think of suitable search words to express search intentions, resulting in existing input associations. The recommendation method has the defects of low efficiency, low accuracy and high input cost

Images