Hierarchical Text Classification And Its Application In News Recommender System

Intelligent information recommendation, because it can be widely applied to various kinds of information system, effectively resolve in the face of vast amounts of information of embarrassment, and establish the adaptive personalized application system, and much attention by researchers and application unit. This study news data of the application as the background, trying to find a kind of effective method of text categorization to establish intelligent recommendation system, can make according to the user’s personal interests to locate and recommend relevant news information.Papers on the analysis of the relevant text classification technology, and recommender system research present situation, especially about feature selection method research and classifier based on the text classification technology, and analyzes the personalized recommendation system is applied to the design and implementation of the case. On this basis, the paper will be based on the content of traditional filtering algorithms and combined with hierarchical text classification, complement each other, make full use of the advantages of content-based filtering, lighten the degree of sparse matrix and influence of early level problem. And hierarchical text classification technology is used to overcome the problem based on content filter can’t be content. To build news recommendation systems, reduce the possibilities in the user lost in the vast network resources, enabling them to quickly and accurately locate you need content, improving the efficiency of the site, to improve the user’s interest.Experimental results prove that this paper the proposed method can indeed produce effective recommendations, test in this paper, several typical recommend other existing methods are compared, and from a hierarchical text classification process in two aspects of work and recommendations for future research are also proposed.