Multi-label informed latent semantic indexing Export

@inproceedings{citeulike:1562553, abstract = {Latent semantic indexing (LSI) is a well-known unsupervised approach for dimensionality reduction in information retrieval. However if the output information (i.e. category labels) is available, it is often beneficial to derive the indexing not only based on the inputs but also on the target values in the training data set. This is of particular importance in applications with multiple labels , in which each document can belong to several categories simultaneously. In this paper we introduce the multi-label informed latent semantic indexing (MLSI) algorithm which preserves the information of inputs and meanwhile captures the correlations between the multiple outputs. The recovered "latent semantics" thus incorporate the human-annotated category information and can be used to greatly improve the prediction accuracy. Empirical study based on two data sets, Reuters-21578 and RCV1, demonstrates very encouraging results.}, address = {New York, NY, USA}, author = {Yu, Kai and Yu, Shipeng and Tresp, Volker}, booktitle = {SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval}, citeulike-article-id = {1562553}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1076034.1076080}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1076034.1076080}, doi = {10.1145/1076034.1076080}, isbn = {1-59593-034-5}, keywords = {categorization, lsi, methods, multi-output, projection, text}, location = {Salvador, Brazil}, pages = {258--265}, posted-at = {2009-09-28 21:30:55}, priority = {2}, publisher = {ACM}, title = {Multi-label informed latent semantic indexing}, url = {http://dx.doi.org/10.1145/1076034.1076080}, year = {2005} }

TY - CONF ID - citeulike:1562553 L3 - citeulike-article-id:1562553 N2 - Latent semantic indexing (LSI) is a well-known unsupervised approach for dimensionality reduction in information retrieval. However if the output information (i.e. category labels) is available, it is often beneficial to derive the indexing not only based on the inputs but also on the target values in the training data set. This is of particular importance in applications with multiple labels , in which each document can belong to several categories simultaneously. In this paper we introduce the multi-label informed latent semantic indexing (MLSI) algorithm which preserves the information of inputs and meanwhile captures the correlations between the multiple outputs. The recovered "latent semantics" thus incorporate the human-annotated category information and can be used to greatly improve the prediction accuracy. Empirical study based on two data sets, Reuters-21578 and RCV1, demonstrates very encouraging results. CY - Salvador, Brazil SN - 1-59593-034-5 AD - New York, NY, USA EP - 265 TI - Multi-label informed latent semantic indexing PB - ACM T2 - SIGIR '05: Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval SP - 258 KW - categorization KW - lsi KW - methods KW - multi-output KW - projection KW - text AU - Yu, Kai AU - Yu, Shipeng AU - Tresp, Volker PY - 2005/// UR - http://dx.doi.org/10.1145/1076034.1076080 ER -