A Method of Constructing a Semantic Network of Interest Subjects Oriented to Social Networks

Abstract

The invention discloses a method for constructing a social network-oriented semantic network of interest topics to solve the problem that the current method of interest perception and message push mainly based on text classification cannot adapt to the growth of new concepts and new points of interest in social networks. Firstly, the required key text information is extracted from the read-in webpage file, and the keyword set is extracted, and then the relationship between keywords in the page is analyzed, and combined with the keyword relationship in the historical page, a keyword semantic network is constructed. Through the subject word discrimination and subject word screening, the similarity calculation of the subject terms is carried out, and the semantic network between the subject terms and the subject terms is determined to be constructed or a new theme is formed. The new word discovery mechanism introduced by the present invention can realize the discovery of new words in social networks, reduce the dictionary configuration operations after new words appear, realize automatic discovery of new interest topics, and effectively provide fresher and more accurate information for retrieval and recommendation services subject categories.

Description

technical field [0001] The invention relates to the field of computer information retrieval, in particular to discovery of topics of interest and construction of semantic links for social network text content, and an automatic update mechanism of the entire topic network. Background technique [0002] With the development of Internet technology and the transformation of Internet users from passive credit granting to active letter sending, social networks and social media gradually dominate the information dissemination of the Internet. A social network is a social network service (Social Network Service). There are various forms of social network, among which blogs, forums and podcasts developed earlier, and gradually produced microblogs, communities, Qzone, WeChat circle of friends, etc. Users write, share, evaluate, discuss, interact and communicate through these social platforms. [0003] With the explosive increase of social network data, users cannot quickly browse th...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a social network-oriented interest topic construction and update method, to solve the problem that the current interest perception and message push method based on text classification cannot adapt to the growth of new concepts and new points of interest in social networks, and realize automatic semi-automatic Topic interest discovery and other functions

Images