Study and Implementation of Accurate Retrieval System Based on Attractions Interest Model

摘要

With the rapid development of Web 2.0 and social networks, travel has become more and more popular. But it becomes very difficult to filter out the travel information which meets the needs from the mass tourism information on the Internet. In this paper, we implement an information retrieval system based on topic model to solve this issue. First, we construct the attractions interest model by using topic model to analyze large attractions comments. Then we use a highly efficient algorithm to implement topic query expansion for realizing semantic query. On this basis, we propose the method of topic distribution sorting. The retrieval system can return more accurate retrieval results to users by speculating user query intent. Experimental results show that the ranking ability of our accurate retrieval system shows advantages beyond traditional retrieval system.