With the advent of the Web, cross-language information retrieval (CLIR) becomes important not only to satisfy the information need across languages but to mine resources for multiple languages e.g. parallel or comparable documents. Broadly CLIR techniques are of two types, in the first case, either queries or documents are translated to the language of comparison while the other type tries to project the vector space representation of the text to a shared translingual space which represents the "semantics" of the documents. In this study, we review the state-of-the-art for CLIR by means of the latter approach and identify the scope for further research.
展开▼
