fit.nb.multinomial: Function fit.nb.multinomial implements model building...

In vovkaOst/TextClassifier: Package for classifying text using Abstracts dataset http://archive.ics.uci.edu/ml/datasets/NSF+Research+Award+Abstracts+1990-2003

Description

Multinomial Naive Bayes is a specialized version of Naive Bayes that is designed more for text documents Multinomial model generates one term from the vocabulary in each position of the document Whereas simple naive Bayes would model a document as the presence and absence of particular words, multinomial naive bayes explicitly models the word counts relevance in particular class / per all words in a class

Usage

fit.nb.multinomial(data, fact)

Arguments

data	data.frame contains values of our training set (bag-of-words format of text)
fact	factor containing proper classification of the training samples in data set (one per observation)

Details

More information about the method can be found http://www.cs.cmu.edu/~knigam/papers/multinomial-aaaiws98.pdf More information about algorithm can be found in the book Introduction to Information Retrieval (p.253-265)

Value

object of the class mcTmnb to use in predict function

Examples

fit.nb.multinomial(train.data, label)

vovkaOst/TextClassifier documentation built on May 3, 2019, 6:41 p.m.