Semantic Relatedness Based Query Translation Disambiguation Approach for Cross-Language Web Search

Abstract

Most of query translation disambiguation approaches rely on a parallel corpus to resolve the translation ambiguity.  This is because parallel corpus are good resources for resolving translation ambiguity. However, they have limitations in translating with a low convergence, specific domain, handling multi-word expressions, and non-availability in all pairs of languages. This work proposes a query translation disambiguation approach, which is based on measuring the semantic relatedness between the terms of the target query. It uses the semantic relatedness to select the suitable translation candidate of query terms and it doesn’t rely on parallel corpus. BabelNet lexical resource is used for extracting translating candidates of words and multi-word expressions. Moreover, splitting of complex compounds, and a back-translation is applied to enhance the final translation. The proposed approach is evaluated using the standard CLEF 2014 dataset collections. The results of the proposed approach are better than the previous corpus-based approaches for the Cross-Language Web Search (CLWS).