Mongolian-Chinese Cross-Language Query Expansion Based on Cross-Language Word Vectors

Abstract

Cross-Language information retrieval is supposed to make information retrieval of one language according to the queries of other languages, and it is a important branch of information retrieval. Recently, cross-language word vectors performs effectively on cross-language information retrieval, and more and more researchers get interested in this field. This paper takes use of cross-language word vectors to map Chinese query words to Mongolian query words, which will be used to perform information retrieval. Three methods were proposed in this paper to perform mapping, namely Series, Series_opt and Cross_valid. These methods are used to map the Chinses queries as well as to select and sort the mapped words. Experimental results conducted in the real environments show that our proposed algorithm can obtain improvement on Mongolian-Chinses cross-language information retrieval.