The Naive Bayes Text Classifier is a variant of the Naive Bayes classification algorithm that is designed specifically for document classification.
For information about the Naive Bayes classification algorithm and functions, see Naive Bayes Functions (ML Engine).
Naive Bayes Text Classifier runs these functions:
- NaiveBayesTextClassifierTrainer2 (ML Engine), which creates a model from training data
- NaiveBayesTextClassifierPredict_MLE (ML Engine), which uses the model to make predictions about testing data
The preceding functions accept input in the form of either documents or tokens. The following figure shows the process if tokenization is performed outside the function. If the input is in the form of documents, both trainer and predictor functions use TextParser (ML Engine) internally to tokenize the data.