A simple php based Naive-Bayes text classifier.
The Naive-Bayes classifier is a statistical classifier which relies on a "trained set" to classify documents.
Documents are chunks of plain text, classes are any type of class structure a document ay fall under.
Examples:
* Restaurant Menu <-> Food Type
* Vocabulary <-> Language
* Novel <-> Author
Before starting to use the classifier you should train it.
This package comes with both a Trainer and a Classifier. To achieve good performance it is recommended to train at least 200 documents before using the classifier.
Note: The 200 document number is an arbitrarily high number. As with any machine learner a golden mean number should be encountered for each specific case.
...
...