Abstract—Vocabulary and Word mismatches are common problems in Information Retrieval Systems. Query Expansion(QE) gives a solution to these problems. Selection of key terms to expand the root query is a challenging task. Semantic relationships of a particular term in WordNet will improve the performance. Pseudo Relevance Feedback is one of the proven methods of QE which consider relevant terms to be expanded. Initial retrieved Items are considered as relevant to the query and further used to expand the query. Whenever the sense of key term is found, its lexical relationship synonym is considered for QE. The specific terms to be included in the initial query are selected automatically and the terms selected for the initial query are connected before expansion with OR Boolean operator. We found there is no Telugu WordNet available hence the proposed system is aimed to test on hand-crafted WordNet on limited text collection for Telugu language.
Index Terms—Query expansion, pseudo relevance feedback, wordnet, synset, word sense disambiguation, parts of speech, normalization.
Ramakrishna Kolikipogu is with Computer Science and Information Technology, Hyderabad (e-mail:Krkrishna.csit@gmail.com).
Padmaja Rani is with a professor & HOD of CSE, Jntuhcoe, Hyderabad.
N. Swapna is with VREC, Hyderabad.
[PDF]
Cite:Ramakrishna Kolikipogu, B. Padmaja Rani, and N. Swapna, "Pseudo Relevance Feedback by linking WordNet for Expanding Queries in Information Retrieval Process," International Journal of Modeling and Optimization vol. 3, no. 5, pp. 462-467, 2013.