A compound model for consumer health search

General search engines are still far from being effective in addressing complex consumer health queries. The language gap between the consumers and the medical resources can confuse non-expert consumers, and may cause problems like the growing concerns about common symptoms. Current methods in addressing this issue are primarily based on modern information retrieval approaches and query expansion is one of the primes. In this paper, an investigation on merging new schemes into state of the art techniques is made and a new compound system based on query expansion approach is presented. This system takes into account the characteristics of medical language and combines Natural Language Processing techniques with traditional query expansion to overcome the query expansion approach shortcomings of not paying enough attention to the specialty of the medical language. The system is evaluated on the CLEF 2017 eHealth IR challenge data and its effectiveness is demonstrated