Bag of words classification project using a Naive Bayes probabilistic model. The program takes a text file to train itself, counting every word associated with the tag and it calculates the probability of an outcome given a word.
To use the program, select a text file with the 'Choose training file' button and with the following structure and hit 'Run':
rainy hot high false | no
rainy hot high true | no
overcast hot high false | yes
sunny mild high false | yes
The left side represents words associated with the right side word.
When the program has calculated all the words probability the results will be available at the left side of the window.
Finally you can enter a phrase and click 'Clasify' to get the most likely outcome with some additional information.
After each classification the program will change it's probability by saving the words of the phrase to the training set. The calculations use Laplace smoothing of a value of 1.