Sentence Retrieval with LSI and Topic Identification

@inproceedings{citeulike:593588, abstract = {This paper presents two sentence retrieval methods. We adopt the task definition done in the TREC Novelty Track: sentence retrieval consists in the extraction of the relevant sentences for a query from a set of relevant documents for that query. We have compared the performance of the Latent Semantic Indexing (LSI) retrieval model against the performance of a topic identification method, also based on Singular Value Decomposition (SVD) but with a different sentence selection method. We used the TREC Novelty Track collections from years 2002 and 2003 for the evaluation. The results of our experiments show that these techniques, particularly sentence retrieval based on topic identification, are valid alternative approaches to other more ad-hoc methods devised for this task.}, address = {London, UK}, author = {Parapar, David and Barreiro, \'{A}lvaro }, booktitle = {Advances in Information Retrieval: 28th European Conference on IR Research, ECIR 2006}, citeulike-article-id = {593588}, keywords = {2get, information-retrieval, web}, month = {April}, pages = {119--130}, posted-at = {2006-04-21 13:26:45}, priority = {3}, title = {Sentence Retrieval with LSI and Topic Identification}, year = {2006} }

TY - CONF ID - citeulike:593588 L3 - citeulike-article-id:593588 TI - Sentence Retrieval with LSI and Topic Identification T2 - Advances in Information Retrieval: 28th European Conference on IR Research, ECIR 2006 SP - 119 EP - 130 AD - London, UK N2 - This paper presents two sentence retrieval methods. We adopt the task definition done in the TREC Novelty Track: sentence retrieval consists in the extraction of the relevant sentences for a query from a set of relevant documents for that query. We have compared the performance of the Latent Semantic Indexing (LSI) retrieval model against the performance of a topic identification method, also based on Singular Value Decomposition (SVD) but with a different sentence selection method. We used the TREC Novelty Track collections from years 2002 and 2003 for the evaluation. The results of our experiments show that these techniques, particularly sentence retrieval based on topic identification, are valid alternative approaches to other more ad-hoc methods devised for this task. KW - 2get KW - information-retrieval KW - web AU - Parapar, David AU - Barreiro, Álvaro PY - 2006/04// ER -