- In this project, I implemented a Naive Bayes text classifier from scratch to categorize the titles of news. The categories are:
- International
- Sport
- Political
- Cultural-artistic
- Social
- Scientific-medical
- Economic
- Social media
- Web browsing
- Video & audio
- All characters except
آ-ی
and\s
have been removed; Numbers have been replaced byN
. Zeros have been handled by Laplacian smoothing. - The
DataPreprocessor
andNaiveBayesClassifier
classes have been designed for preprocessing the samples and training/evaluating a Naive Bayes classifier respectively.
mohammad8921 / newstitlescategorization-naivebayesclassifier Goto Github PK
View Code? Open in Web Editor NEWA Naive-Bayes classifier to categorize titles of persian news