Personalized webpage recommendation method based on topic and relative entropy

Abstract

The present invention discloses a personalized webpage recommendation method based on a topic and a relative entropy. According to the method, firstly, an LDA (latent dirichlet allocation) model is adopted to carry out topic mining on webpage content and user reading behaviors and to calculate a webpage semantic feature vector and a user interest feature vector based on the topic; and then a similarity measuring formula based on the concept of the relative entropy is utilized to calculate similarity between a webpage-to-be-recommended semantic feature vector and the user interest feature vector, and the obtained similarity is used as a decision basis for personalized webpage recommendation. According to the personalized webpage recommendation method based on the topic, a great deal of computing cost based on a collaborative filtering method is avoided; and meanwhile, the topic, instead of a keyword, is adopted to represent webpage content, and thus, the recommendation process and the recommendation results can more comprehensively and accurately reflect conceal information and deep semantic features of the webpage content.

Description

technical field [0001] The invention relates to a webpage personalized recommendation method based on topics and relative entropy, which can be used in network applications such as user interest identification, webpage personalized recommendation, and news push on demand, and belongs to the technical field of the Internet. Background technique [0002] With the rapid development of the Internet and the continuous enrichment of online information resources, the World Wide Web (abbreviated as the Web) has become the most important place for people to obtain information, understand news and current events, and pursue interesting content. However, the massive web page information resources in the Web often reflect the characteristics of dynamics, non-structurality, and disorder. Most public websites collect a large number of web pages according to popular needs. Different users see the same content organization, resulting in poor user experience. poor. At the same time, because...

AI Technical Summary

Problems solved by technology

However, the personalized recommendation method based on collaborative filtering has problems such as sparsity and scalability, and it requires a large computational overhead, so it is often difficult to adapt to mobile scenarios with limited computing power of devices and frequent changes in user groups.

The content-based personalized recommendation method mostly uses the TF-IDF algorithm to directly extract the keywords of the web content, but the superficial features of the keywords are often difficult to fully reflect the deep semantics contained in the content, and due to the lack of a unified keyword mapping standard , which often greatly increases the difficulty of similarity calculation and recommendation decision-making

Images