Tolerant Retrieval and Query Processing in Search Engine

Tolerant retrieval and query processing in search engine can enhance the robustness and users feel convenient. Based on the similar comparison between the right terms and terms proposed by users, the proposed approach can not only tolerant misspelling, but also present relevant retrieval advice. If more than two correctly spelled terms are nearly tied, it can also select the one that is more common. The proposed algorithms are useful in the case when user may not know how a proper term is spelled. Experimental results and the analysis show the feasible of the approach, and existing problems and further works are also present in the end.