Assessing Query Translation Quality Using Back Translation in Hindi-English CLIR
Ganesh Chandra, Sanjay Kumar Dwivedi · International Journal of Intelligent Systems and Applications · 2017
Cross-Language Information Retrieval (CLIR) is a most demanding research area of Information Retrieval (IR) which deals with retrieval of documents different from query language.In CLIR, translation is an important activity for retrieving relevant results.Its goal is to translate query or document from one language into another language.The correct translation of the query is an essential task of CLIR because incorrect translation may affect the relevancy of retrieved results.The purpose of this paper is to compute the accuracy of query translation using the back translation for a Hindi-English CLIR system.For experimental analysis, we used FIRE-2011 dataset to select Hindi queries.Our analysis shows that back translation can be effective in improving the accuracy of query translation of the three translators used for analysis (i.e.Google, Microsoft and Babylon).Google is found best for the purpose.