Font Size: a A A

Design And Implementation The Personalized Recommendation System Based On Hot Network

Posted on:2015-03-23Degree:MasterType:Thesis
Country:ChinaCandidate:X X WangFull Text:PDF
GTID:2268330428969442Subject:Software engineering
Abstract/Summary:PDF Full Text Request
The rapid development of Internet make us in the era of information overload, how to obtain the user’s attention and the most interesting information has become an urgent problem to be solved from these massive information.The search engine appears to solve the user search information needs, if the keywords users cannot describe the needed information, is not retrieved, we need a more intelligent system to satisfy these requirements, in this context, recommendation system emerge as the times require.However, the existing recommender systems and are not taken into account in particular,intelligence applications such as heat, information, timely issues, eventually led to the poor, on information recommendation effect to this end, this paper presents a recommendation technology, the hot spot of network intelligence based on the other, in order to solve a problem of fitting and user cold start the user interest problems, this paper interest modeling for users in recommender systems, an accurate understanding of the user’s current environment demand, then, based on the singular value decomposition of the collaborative filtering algorithm is the key technology to design a recommendation system based on effective, to meet the needs of users of information. Finally, the design and implementation of the personalized information recommendation system.Firstly, this paper describes the application of the significance of the research, information recommendation system of foreign research status and research content and recommendation system in practice.Secondly, recommendation system demand analysis foundation in personalized information, this paper designs the overall architecture and function of information recommendation system modules, each module of the detailed design and the code realization, and the key techniques used in the system are described in detail. Finally, the personalized information network hot recommendation system is tested based on the experimental results, on the basis of the analysis, put forward the need to further improve and improvement aspects.
Keywords/Search Tags:Information recommendation, hot spots, collaborative filtering, Interest modeling
PDF Full Text Request
Related items