User identity-based input association recommendation method and apparatus

Abstract

The invention provides a user identity-based input association recommendation method. The method comprises the steps of identifying a user identity according to a search scene of a user, and trackinga search behavior log corresponding to the user identity; adding the search behavior log to an existing input association candidate model, and generating input association candidate word banks after fusion; receiving a search query input by the user, and sorting the input association candidate word banks after fusion in combination with the search query; and recommending the sorted input association candidate word banks as a recommendation result of the search query. Therefore, the user identity can be uniquely identified in a complex network environment, a search clicking behavior of the useris stably and continuously recorded, and a foundation is laid for subsequently providing recommendation information for the user pointedly. According to the method, the required information can be recommended for the user more accurately; the exporting and importing of input association are greatly improved; the search cost of the user is reduced; and the search experience of the user is improved.

Description

technical field [0001] The invention relates to the technical field of computer networks, in particular to an input association recommendation method and device based on user identity. Background technique [0002] With the continuous development of computer network technology, the network is flooded with various information. At present, the method of obtaining information through the network is not only fast in information acquisition speed, but also obtains a large amount of information and rich content, which is easier to meet user needs. Therefore, in today's society, more and more users search and obtain information through the Internet. [0003] At this stage, in order to further improve the user's search speed, an input association method is proposed, that is, after the user enters a certain query through the search engine, candidate words are automatically recommended under the search box, and the user is recommended to the user. other inquiries. At this point, th...

AI Technical Summary

Problems solved by technology

However, the input association method in the prior art is recommended for global users, that is, when all users input specific query words through the network, the recommended words obtained are consistent

However, there are differences among individual users, and different users have different search habits. The existing input association methods cannot provide specific users with recommended information that meets their own actual needs, reducing the accuracy of recommended information.

Images