Latent semantic models for collaborative filtering Export

@article{citeulike:344216, abstract = {Collaborative filtering aims at learning predictive models of user preferences, interests or behavior from community data, that is, a database of available user preferences. In this article, we describe a new family of model-based algorithms designed for this task. These algorithms rely on a statistical modelling technique that introduces latent class variables in a mixture model setting to discover user communities and prototypical interest profiles. We investigate several variations to deal with discrete and continuous response variables as well as with different objective functions. The main advantages of this technique over standard memory-based methods are higher accuracy, constant time prediction, and an explicit and compact model representation. The latter can also be used to mine for user communitites. The experimental evaluation shows that substantial improvements in accucracy over existing methods and published results can be obtained.}, address = {New York, NY, USA}, author = {Hofmann, Thomas}, citeulike-article-id = {344216}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=963774}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/963770.963774}, doi = {10.1145/963770.963774}, issn = {1046-8188}, journal = {ACM Trans. Inf. Syst.}, keywords = {collabfilter, topicinference}, month = {January}, number = {1}, pages = {89--115}, posted-at = {2005-12-06 14:57:41}, priority = {3}, publisher = {ACM}, title = {Latent semantic models for collaborative filtering}, url = {http://dx.doi.org/10.1145/963770.963774}, volume = {22}, year = {2004} }

TY - JOUR ID - citeulike:344216 L3 - citeulike-article-id:344216 N2 - Collaborative filtering aims at learning predictive models of user preferences, interests or behavior from community data, that is, a database of available user preferences. In this article, we describe a new family of model-based algorithms designed for this task. These algorithms rely on a statistical modelling technique that introduces latent class variables in a mixture model setting to discover user communities and prototypical interest profiles. We investigate several variations to deal with discrete and continuous response variables as well as with different objective functions. The main advantages of this technique over standard memory-based methods are higher accuracy, constant time prediction, and an explicit and compact model representation. The latter can also be used to mine for user communitites. The experimental evaluation shows that substantial improvements in accucracy over existing methods and published results can be obtained. IS - 1 JF - ACM Trans. Inf. Syst. SN - 1046-8188 AD - New York, NY, USA EP - 115 TI - Latent semantic models for collaborative filtering PB - ACM VL - 22 SP - 89 KW - collabfilter KW - topicinference AU - Hofmann, Thomas PY - 2004/01// UR - http://dx.doi.org/10.1145/963770.963774 ER -