Automatic Named Entity Set Expansion Using Semantic Rules and Wrappers for Unary Relations

Named Entity Set Expansion is the task from a set given a small number of named entities in same type finds a more complete Named Entity set. Two famous Named Entity expansion systems using the web are Google Sets and Boo!Wa!. In this paper, we proposed an approach using Vietnamese semantic rule combined with wrappers for unary relations to expand a given set of Vietnamese Named Entities. Our model using the results which are provided by Google search engine in Vietnam, consists of six main phases. We did the experiments for many times to choose the rank method that gave the best results is Page Rank, and mixing coefficient is 0.6. Experimental results showed that the feasible approach with average precision is 96.4% with 5 first entities and 92.6% with 10 first entities.