| URL: | http://sourceforge.net/projects/sgmweka/ | |
| Author: | Antti Puurula<asp12{[at]}students.waikato.ac.nz> | |
| Maintainer: | Antti Puurula<asp12{[at]}students.waikato.ac.nz> |
A wrapper for the SGM toolkit for generative modeling with sparse matrix representations. Implements Multinomial Naive Bayes and Multinomial Kernel Density classifiers, and modifications such as TF-IDF transforms, different types of smoothing, pruning and model based feedback. Default classifier is a Multinomial Naive Bayes model with TF-IDF term weighting and smoothing with a uniform background model. Restart Weka after package installation. Multi-label classification, ranked retrieval and other functions are available through both command-line calls and class inclusion.