Relevance based language models Export

@inproceedings{citeulike:912092, abstract = {We explore the relation between classical probabilistic models of information retrieval and the emerging language modeling approaches. It has long been recognized that the primary obstacle to effective performance of classical models is the need to estimate a relevance model : probabilities of words in the relevant class. We propose a novel technique for estimating these probabilities using the query alone. We demonstrate that our technique can produce highly accurate relevance models, addressing important notions of synonymy and polysemy. Our experiments show relevance models outperforming baseline language modeling systems on TREC retrieval and TDT tracking tasks. The main contribution of this work is an effective formal method for estimating a relevance model with no training data.}, address = {New York, NY, USA}, author = {Lavrenko, Victor and Croft, W. Bruce}, booktitle = {SIGIR '01: Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval}, citeulike-article-id = {912092}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=383972}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/383952.383972}, doi = {10.1145/383952.383972}, isbn = {1-58113-331-6}, keywords = {expansion, feedback, ir, prf, pseudo, query, relevance}, location = {New Orleans, Louisiana, United States}, pages = {120--127}, posted-at = {2007-11-29 21:37:49}, priority = {0}, publisher = {ACM}, title = {Relevance based language models}, url = {http://dx.doi.org/10.1145/383952.383972}, year = {2001} }

TY - CONF ID - citeulike:912092 L3 - citeulike-article-id:912092 N2 - We explore the relation between classical probabilistic models of information retrieval and the emerging language modeling approaches. It has long been recognized that the primary obstacle to effective performance of classical models is the need to estimate a relevance model : probabilities of words in the relevant class. We propose a novel technique for estimating these probabilities using the query alone. We demonstrate that our technique can produce highly accurate relevance models, addressing important notions of synonymy and polysemy. Our experiments show relevance models outperforming baseline language modeling systems on TREC retrieval and TDT tracking tasks. The main contribution of this work is an effective formal method for estimating a relevance model with no training data. CY - New Orleans, Louisiana, United States SN - 1-58113-331-6 AD - New York, NY, USA EP - 127 TI - Relevance based language models PB - ACM T2 - SIGIR '01: Proceedings of the 24th annual international ACM SIGIR conference on Research and development in information retrieval SP - 120 KW - expansion KW - feedback KW - ir KW - prf KW - pseudo KW - query KW - relevance AU - Lavrenko, Victor AU - Croft, Bruce PY - 2001/// UR - http://dx.doi.org/10.1145/383952.383972 ER -