A Semantic Search System For Traditional Chinese Medicine

With the rapid development of internet and information technology widely applied in various science fields, data is organized in various schemas for different organizations in the same field. How to share and exploit these data resources efficiently in various organizations is a big problem to be solved. For this goal, we presented a new distributed semantic search system based on DartSearchV3 in our lab. This paper describes the details of the design and implementation of this system.At first, we briefly introduced the latest development of search engine and trouble of traditional search engines. And then the related work such as DartGrid and DartSearchV3 in our lab was introduced. Then, we described the open source tools Hadoop and Lucene followed in this distributed semantic search system. At last, according to the new demands of data integration of Traditional Chinese Medicine and the defects of old DartSearchV3, we presented the distributed semantic search system and introduced the details of the design and implementation.This paper focused on the system design of the distributed semantic search system, the algorithm of hybrid semantic index, how to integrate multiple distributed data centers semantically and how to make distributed index and search in Hadoop. Furthermore, this paper introduced the process of hybrid semantic index and the management of distributed indexes and the dispatch of the search tasks in Hadoop.In the last part of this thesis, several directions are pointed out in the future work for this distributed semantic search system.