Hiemstra, D.: Term-Specific Smoothing for the Language Modeling Approach to Information Retrieval. Hiemstra, D.: Using Language Models for Information Retrieval. ACM Transactions on Information Systems 20, 357–389 (2002) Information Processing & Management 36, 95–108 (2002)Īmati, G., van Rijsbergen, C.J.: Probabilistic Models of Information Retrieval Based on Measuring the Divergence from Randomness. Robertson, S.E., Walker, S., Beaulieu, M.: Experimentation as a Way of Life: Okapi at TREC. Experiment and Evaluation in Information Retrieval. ACM - Transactions on Asian Languages Information Processing 4, 163–189 (2005)ĭolamic, L., Savoy, J.: UniNE at FIRE 2008: Hindi, Marathi and Bengali IR. Savoy, J.: Comparative Study of Monolingual and Multilingual Search Models for Use with Asian Languages. Savoy, J.: Combining Multiple Strategies for Effective Monolingual and Cross-Lingual Retrieval.
Applying the Z-score as data fusion operator after a blind-query expansion tends also to improve the MAP of the merged run over the best single IR system. For these languages, our various experiments demonstrate also that either an aggressive stemming procedure or the trunc- n indexing approach produces better retrieval effectiveness when compared to other word-based or n-gram language-independent approaches. To evaluate these solutions we have used various IR models including models derived from Divergence from Randomness (DFR), Language Model (LM) as well as Okapi, or the classical tf idf vector-processing approach.įor the three studied languages, our experiments tend to show that IR models derived from Divergence from Randomness (DFR) paradigm tend to produce the best overall results.
As another language-independent indexing strategy, we have evaluated the trunc- n method in which the indexing term is formed by considering only the first n letters of each word. We have compared their retrieval effectiveness with both light stemming strategy and n-gram language-independent approach. As a second goal, we have developed new and more aggressive stemming strategies for both Marathi and Hindi languages during this second campaign. Our first objective in participating in FIRE evaluation campaigns is to analyze the retrieval effectiveness of various indexing and search strategies when dealing with corpora written in Hindi, Bengali and Marathi languages.
0 kommentar(er)
