Calculating Similarity between Unknown Words Based on Combination Strategy
Fan Xing-Hua, Cao Rong-Li · Applied Mathematics & Information Sciences · 2013
Abstract: This paper presents a method of calculating similarity between unknown words based on combination strategy. In the method, the best concept expression of an unknown word in corpus is obtained from the background of it, and then we construct context for the best concept expression. The connotation meaning of an unknown word is determined by the difference between the context of the best concept expression and its own context. The similarity between unknown words is calculated by utilizing semantic dictionary and the method of using the search engine to reconstruct corpus. Experimental results show that the precision, the recall and F1 in the method reaches 99.4%, 97.3 % and 98.4 % respectively, which are increased 21.4%, 3.2 % and 13.1 % respectively compared with the results got by using the method of similarity calculation based on HowNet known.