Searching method and device applying label knowledge network

Abstract

The invention discloses a searching method and device applying a label knowledge network. The method comprises the following steps: obtaining a plurality of recommended articles, carrying out label extraction on text information related to each recommended article to obtain one or more corresponding article labels, and determining an article label set formed by all article labels; determining historical behavior data of the user for different articles, analyzing user tags preferred by the user according to the historical behavior data, and determining a user tag set composed of all the user tags; constructing a label knowledge network through the article label set, the knowledge graph and the Word2vec model; generating an article feature vector of the recommended article and a user featurevector of the user according to the article tag set, the user tag set and the tag knowledge network; respectively performing retrieving to obtain related articles or related users through the articlefeature vectors or the user feature vectors. In this way, correlation of recall results of the content algorithm can be ensured, and the defects of semantic limitation and poor expansibility of tag recall results can be overcome.

Description

technical field [0001] The present application relates to the technical field of intelligent search, in particular, to a search method and device using a tag knowledge network. Background technique [0002] With the development of Internet technology and social networks, a large amount of information including text information, pictures, videos, etc. are published on the Internet every day. Traditional search technology has been unable to meet the needs of users for information discovery, and personalized recommendation system came into being to solve the problem of information overload. It can recommend the required information to users according to their interests and behaviors, and help users quickly discover and satisfy massive amounts of information, so as to improve users' stickiness to products, improve user retention, and improve product competitiveness. [0003] Currently commonly used recommendation algorithms include content-based recall algorithms, collaborati...

AI Technical Summary

Problems solved by technology

[0008] 2. The mining accuracy of similar users and similar items is poor

[0009] Recommendation systems rarely use tags to mine similar users and items, mainly because the tags are too fine and the scalability is poor

The labels of items are generally generated by machines based on the text information of the items, because in the face of massive items, the cost of manual labeling is too high

The simple model does not have rich prior knowledge like knowledge experts, and cannot distinguish whether "Andy Lau" and "Andy Lau" are labels with the same semantic meaning

[0010] Aiming at the semantic limitations of recall results, poor scalability, and poor mining accuracy of similar users and similar items in related technologies, no effective solutions have been proposed so far.

Images