Support The Research And Design Personalized Recommendation Information Mining System

With the rapid development of the Internet makes a lot of information at the same time in front of us, a massive source of information can provide rich sources of intelligence information and intelligence more and more attention of the intelligence officers, and Internet applications can not be love interest for different users provide appropriate services. The explosion of information makes the utilization of information but reduced, this phenomenon is called information overload. To solve this problem, we will produce a personalized recommendation technology.Current personalized recommendation technology research summary, the first details of existing content-based information filtering technology and in-depth understanding of collaborative filtering technology, the sparsity problem in collaborative filtering algorithm cold start problem and the existing solutions. Based on in-depth analysis of these two methods, a new hybrid recommendation algorithm based on user behavior and user-intelligence personalized recommended. The algorithm recommended on the basis of the user based on similar users and intelligence combined using collaborative filtering algorithm based on the user-intelligence ". The main idea is manifested from user behavior to predict similar users attention of the attention of the intelligence content Finally Recommended intelligence.Finally, the use of personalized recommendation algorithm complete support personalized recommendation intelligence mining system design and implementation.