Obtained some machine learning data for classification, tested various SKLearn implementations of Naïve Bayes, and reported the findings.
This is a python program where we are reading the data from the Reuters Corpus 21578 new articles. Here we are reading the Topics, DateLine and topics of each article. We split the dataset into 80/20 where first 80% of the dataset is used to train the algorithm and other 20% of the dataset is used to test the algorithm. We have used same dataset split for each run and each algorithm to compare the performance using confusion matrix, Precision and Recall.
