@techreport{Eyh03, title = {Sparse Bayesian Classifiers for Text Categorization}, author = {S. Eyheramendy and A. Genkin and W.H. Ju and D.D. Lewis and D. Madigan}, institution = {DIMACS Working Group on Monitoring Message Streams}, year = {2003}, biburl = {http://www.bibsonomy.org/bibtex/26b6e6b547f415c09456af7e533216b91/mcdiaz}, file = {#F#}, keywords = {imported } } @book{genkin2003, title = {Organizacija, normirovanie i oplata truda na promyÿéslennych predprijatijach : uÿécebnik dlja vuzov}, address = {NORMA}, annote = {X, 389 S.}, author = {{Boris Michajloviÿéc} Genkin}, howpublished = {Moskva}, url = {http://gso.gbv.de/DB=2.1/CMD?ACT=SRCHA&SRT=YOP&IKT=1016&TRM=ppn+548301387&sourceid=fbw_bibsonomy}, year = {2003}, biburl = {http://www.bibsonomy.org/bibtex/227b3e6037eff0d946921e0e905db73fa/fbw}, description = {imported}, isbn = {5-89123-665-6}, keywords = {imported } } @article{Genkin:August2007:0040-1706:291, title = {Large-Scale Bayesian Logistic Regression for Text Categorization}, author = {Alexander Genkin and David D. Lewis and David Madigan}, journal = {Technometrics}, pages = {291-304(14)}, url = {http://www.ingentaconnect.com/content/asa/tech/2007/00000049/00000003/art00007}, volume = {49}, year = {August 2007}, biburl = {http://www.bibsonomy.org/bibtex/211830035a44e5db49d194a7f7a3f35ed/jhammerb}, description = {IngentaConnect Large-Scale Bayesian Logistic Regression for Text Categorization}, abstract = {Logistic regression analysis of high-dimensional data, such as natural language text, poses computational and statistical challenges. Maximum likelihood estimation often fails in these applications. We present a simple Bayesian logistic regression approach that uses a Laplace prior to avoid overfitting and produces sparse predictive models for text data. We apply this approach to a range of document classification problems and show that it produces compact predictive models at least as effective as those produced by support vector machine classifiers or ridge logistic regression combined with feature selection. We describe our model fitting algorithm, our open source implementations (BBR and BMR), and experimental results.}, doi = {doi:10.1198/004017007000000245}, keywords = {classification data_mining logistic regression text_mining } } @inproceedings{conf/sigir/DayanikLMMG06, title = {Constructing informative prior distributions from domain knowledge in text classification.}, author = {Aynur A. Dayanik and David D. Lewis and David Madigan and Vladimir Menkov and Alexander Genkin}, booktitle = {SIGIR}, crossref = {conf/sigir/2006}, editor = {Efthimis N. Efthimiadis and Susan T. Dumais and David Hawking and Kalervo Järvelin}, pages = {493-500}, publisher = {ACM}, url = {http://dblp.uni-trier.de/db/conf/sigir/sigir2006.html#DayanikLMMG06}, year = {2006}, biburl = {http://www.bibsonomy.org/bibtex/2635997d34b77c2dde18c229867257017/flawed}, abstract = { Supervised learning approaches to text classification are in practice often required to work with small and unsystematically collected training sets. The alternative to supervised learning is usually viewed to be building classifiers by hand, using a domain expert's understanding of which features of the text are related to the class of interest. This is expensive, requires a degree of sophistication about linguistics and classification, and makes it difficult to use combinations of weak predictors. We propose instead combining domain knowledge with training examples in a Bayesian framework. Domain knowledge is used to specify a prior distribution for the parameters of a logistic regression model, and labeled training data is used to produce a posterior distribution, whose mode we take as the final classifier. We show on three text categorization data sets that this approach can rescue what would otherwise be disastrously bad training situations, producing much more effective classifiers.}, ee = {http://doi.acm.org/10.1145/1148170.1148255}, isbn = {1-59593-369-7}, date = {2006-08-30}, keywords = {classification priors } } @inproceedings{conf/sigir/DayanikLMMG06, title = {Constructing informative prior distributions from domain knowledge in text classification.}, author = {Aynur A. Dayanik and David D. Lewis and David Madigan and Vladimir Menkov and Alexander Genkin}, booktitle = {SIGIR}, crossref = {conf/sigir/2006}, editor = {Efthimis N. Efthimiadis and Susan T. Dumais and David Hawking and Kalervo Järvelin}, pages = {493-500}, publisher = {ACM}, url = {http://dblp.uni-trier.de/db/conf/sigir/sigir2006.html#DayanikLMMG06}, year = {2006}, biburl = {http://www.bibsonomy.org/bibtex/2635997d34b77c2dde18c229867257017/dblp}, description = {dblp}, ee = {http://doi.acm.org/10.1145/1148170.1148255}, isbn = {1-59593-369-7}, date = {2006-08-30}, keywords = {dblp } } @article{journals/cad/ArtobolevskiGSGS75, title = {The method of LP - search for the optimization of multiparametric and multicriterial problems in engineering design.}, author = {I. I. Artobolevski and W. M. Grinkewitch and I. M. Sobol and M. D. Genkin and W. I. Sergeyev and R. B. Statnikov}, journal = {Computer-Aided Design}, number = {3}, pages = {160-162}, url = {http://dblp.uni-trier.de/db/journals/cad/cad7.html#ArtobolevskiGSGS75}, volume = {7}, year = {1975}, biburl = {http://www.bibsonomy.org/bibtex/2dbf75369e074de7c58c4b3d958de8fa5/dblp}, description = {dblp}, ee = {http://dx.doi.org/10.1016/0010-4485(75)90005-6}, date = {2004-01-22}, keywords = {dblp } } @article{journals/jifs/GenkinKM02, title = {Set covering submodular maximization: An optimal algorithm for data mining in bioinformatics and medical informatics.}, author = {Alexander Genkin and Casimir A. Kulikowski and Ilya B. Muchnik}, journal = {Journal of Intelligent and Fuzzy Systems}, number = {1}, pages = {5-17}, url = {http://dblp.uni-trier.de/db/journals/jifs/jifs12.html#GenkinKM02}, volume = {12}, year = {2002}, biburl = {http://www.bibsonomy.org/bibtex/2107a79de145242a3abf097f78f542450/dblp}, description = {dblp}, ee = {http://iospress.metapress.com/openurl.asp?genre=article&issn=1064-1246&volume=12&issue=1&spage=5}, date = {2003-08-20}, keywords = {dblp } }