Information Retrieval Based on Telugu Cross-Language Transliteration

Link to full Paper

In the current scenario, information access over cyberspace has increased gradually. A huge amount of information in different languages are available on cyberspace. Most of the people want to retrieve the information in their native languages. This paper portrays regarding an English-Telugu cross-lingual information retrieval (CLIR) system. CLIR fetches Telugu documents to answer a specified English (or) Telugu query. In Indian subcontinent, Telugu is the most well-known spoken dialect in the states of Telangana, Andhra Pradesh and Union Territories of Andaman Nicobar and Yanam. A cross-lingual information retrieval process desires a parallel corpora or bilingual dictionary, which is used to translate texts from one language to another. For in-house translation, the proposed approach uses a bilingual dictionary. This scheme also supports the monolingual information retrieval method. This study also presents the effect of using language dependent and independent stemmers in monolingual information retrieval system.

Nifty tech tag lists fromĀ Wouter Beeftink