BiLingual Information Retrieval System for English and Tamil

TL;DR

This paper presents a bilingual information retrieval system for English and Tamil focused on festivals, utilizing ontological trees and POS tagging to dynamically process and retrieve relevant information in the user's query language.

Contribution

It introduces a generic, extendable bilingual IR platform that dynamically handles queries in English and Tamil using ontological trees and POS tagging.

Findings

Effective retrieval of festival information in English and Tamil

Dynamic language processing improves user query handling

Ontology-based approach enhances multilingual information extraction

Abstract

This paper addresses the design and implementation of BiLingual Information Retrieval system on the domain, Festivals. A generic platform is built for BiLingual Information retrieval which can be extended to any foreign or Indian language working with the same efficiency. Search for the solution of the query is not done in a specific predefined set of standard languages but is chosen dynamically on processing the user's query. This paper deals with Indian language Tamil apart from English. The task is to retrieve the solution for the user given query in the same language as that of the query. In this process, a Ontological tree is built for the domain in such a way that there are entries in the above listed two languages in every node of the tree. A Part-Of-Speech (POS) Tagger is used to determine the keywords from the given query. Based on the context, the keywords are translated to appropriate languages using the Ontological tree. A search is performed and documents are retrieved based on the keywords. With the use of the Ontological tree, Information Extraction is done. Finally, the solution for the query is translated back to the query language (if necessary) and produced to the user.