Kannada and Telugu Native Languages to English Cross Language Information Retrieval
V. Reddy. Mallamma, M. Hanumanthappa · ePrints@Bangalore University (Bangalore University) · 2011
One of the crucial challenges in cross lingual information retrieval is the retrieval of relevant information for a query expressed in as native language. While retrieval of relevant documents is slightly easier, analysing the relevance of the retrieved documents and the presentation of the results to the users are non-trivial tasks. To accomplish the above task, we present our Kannada English and Telugu English CLIR systems as part of Ad-Hoc Bilingual task. We take a query translation based approach using bi-lingual dictionaries. When a query words not found in the dictionary then the words are transliterated using a simple rule based approach which utilizes the corpus to return the ‘k’ closest English transliterations of the given Kannada/Telugu word. The resulting multiple translation/transliteration choices for each query word are disambiguated using an iterative page-rank style algorithm which, based on term-term co-occurrence statistics, produces the final translated query. Finally we conduct experiments on these translated query using a Kannada/Telugu document collection and a set of English queries to report the improvements, performance achieved for each task is to be presented and statistical analysis of these results are given.