A Standard Bibliography Recommended Method Based on Topic Model and Fusion of Multi-feature

Abstract

This paper proposed a recommended method of standard bibliography based on topic model which fused multi-feature. Firstly, the LDA topic model was used to analyze the standard resource which user concerned, then the user attention model was created by combined with the user's information, Secondly, by analyze the feature of standard bibliography documents in attribute, classification and association relationship, the semi-supervised graph clustering algorithm was proposed to realize the construction of the standard bibliography topic model, Finally, the standard bibliography model and user attention model were used to complete the calculation of similarity, by using Top-N algorithm, the highest standard resource was recommend to users. Some experiments based on the Standard Library have been made, the results shown that the F value in the method which proposed in this paper is about 9% higher than the recommendation algorithm based on vector space model, and about 5% higher than the recommended method based on implicit topic model.