A Discourse-Based Information Retrieval for Tamil Literary Texts

Tamil literature has many valuable thoughts that can help the human community to lead a successful and happy life. Tamil literary works are abundantly available and searchable on the Internet. However, the existing search systems follow a keyword-based match strategy that fails to satisfy user needs. This necessitates the demand for a focused information retrieval system that semantically analyzes the Tamil literary text, which will eventually improve the search system performance. This paper proposes a novel information retrieval framework that uses discourse processing techniques in aiding semantic analysis and representation of Tamil literary texts. The proposed framework was tested using two ancient literary works, the Thirukkural and Naladiyar, which were written in 300 Before the Common Era (BCE). The Thirukkural comprises 1,330 couplets, each 7 words long, while the Naladiyar consists of 400 quatrains, each 15 words long. The proposed system, tested with all the 1,330 Thirukkural couplets and 400 Naladiyar quatrains, achieved a mean average precision (MAP) score of 89 percent. The performance of the proposed framework was compared with Google Tamil search and a keyword-based search, which was a substandard version of the proposed framework. Google Tamil search achieved a MAP score of 56 percent while keyword-based method achieved a MAP score of 62 percent. It showed that the discourse processing techniques could improve the search performance of an information retrieval system.