Design and implementation of personalized search model on user's interest

摘要

The traditional information retrieval technology can not satisfy the personalized queries. To solve this problem, we have researched and improved the standard BP algorithm to find user's potential demands, and finally an improved BP neural network algorithm based on personalized search method is proposed. The experiment shows that the algorithm can effectively enhance BP algorithm convergent speed and avoid falling into local minimum, and significantly improve the recommendation quality of the personalization system.