Research on Application of Lucene Search Engine in Social Network Platform
Mei Yu,Wentao Xing,Jian Yu,Jie Gao,Shengguang Ma,Tenghai Wang
DOI: https://doi.org/10.14257/ijhit.2017.10.8.04
2017-01-01
Abstract:With the development of Web2.0 technology, social networks begin to play an increasingly important role in people's life.Widely used social network for researchers has brought some potentially useful information, such as the user's interests and preferences.At the same time, constant search results provided by the search engine can not meet the individual needs of users, a new way search engine, personalized search is an urgent need to explore.Based on this demand, the paper from Sina microblog mining user interests, and the use of open-source Lucene search engine completes personalized search.In this paper, the structure and principle of Lucene search engine are summarized and some related knowledge are introduced, such as text preprocessing and vector space model.Then, this article proposes the Lucene TagMatch Ranking (LTR) algorithm.The main idea is using user's Sina microblog texts to extract tags of interest and measure the matching degree between web pages and user's interests which named value of tag matching degree by the vector space model, then combine the traditional Lucene scoring mechanism, finally realize personalized ranking results based on the user's interest.At last the Eclipse programming algorithm based on Java is used to carry out comparison experiment to confirm the effectiveness of the algorithm.The results are presented to the user in the ranking which based on the user's interests, it will be able to reach recommendation algorithm based on user's interests.