A Personalized Recommendation System Based On Vector Space Model

Along with the constant development of the Internet, information overload fans to their own resources and people on the efficient use of the Internet has become the information bottleneck. It is hoped that the content of the website, to the extent possible,to be adjusted according to the user's browsing interest, so that every user feel like they are the only website users. The key to achieving this goal is to discover how Web users Preferences, Dynamically customized for viewing the contents of the proposed visit or provide. This is defined as the technique of web Personalized recommendation. It is a central issue in the research and application of web techniques.Intending for the specific application requirement on the personalized recommendation for online web lyceum, the paper probes deeply into the meta-search engine and Vector Space Model:1) Meta-search engine. Meta-search engine is a search engine based on a search engine, also known as"the mother of search engines". By the Meta search engine is responsible for conversion processes and submitted to multiple pre-selected independent search engine, and the independent search engine returns all the query results to focus on and then returned to the user.2) Similarity judgmental algorithm -----Vector space model. The biggest advantage of vector space model is it's a huge advantage in knowledge representation method. Vector space model presented in accordance with the query words and documents separately to quantify the dimensions of key words, and then by calculating the cosine angle between two vectors means to be a document with the query word similarity. Thus those who give priority to search and query large documents word similarity, and to be able to retrieve documents according to similarity with the query words to sort. Vector space model is based on text processing a variety of applications to the foundation and prerequisite for achieving. To judge your personality through the information and the similarity search of the resources to conduct personalized recommendation, the recommendation of the recommendation system to improve the quality and accuracy of recommendation.Aiming at the specialty that the Web Log has a high dimension and a huge data,we use similarity algorithm to determine log files for online Academy of user browsing patterns mining. Experimental results show that by the user logs in the user characteristics of items and Web learning resources similarity judgments, web log mining can be combined with a preliminary meet personalized online resources recommended.This paper designs and implements a part of the vector space model based on personalized recommendation system. The system consists of three modules: data preprocessing module, similarity judgments modules, and personalized recommendation Module. The actual use show that the personalized recommendation system files from the Web Log Mining interested in browsing mode the user to provide users with better referral services, guiding users to browse, improve search efficiency and accuracy. This system experiments on the Academy Web site, the basic realization of the envisioned functionality, this system of research and design in online learning field of study has a certain academic value and application value.