Cross Lingual Lexical Substitution Using Word Representation in Vector Space

Polysemous words acquire different senses and meanings from their contexts. Representing words in vector space as a function of their contexts captures some semantic and syntactic features for words and introduces new useful relations between them. In this paper, we exploit different vectorized representations for words to solve the problem of Cross Lingual Lexical Substitution. We compare our techniques with different systems using two measures: “best” and “out-of-ten” (oot), and show that our techniques outperform the state of the art in the “oot” measure while keeping a reasonable performance in the “best” measure.