This project is about Arabic document classification. The documents contained three classes. And ISRI Stemmer was used to stem the text. 6000 texts were used to build the bag of words with respective classes and features used as the number of occurrences. Three classifiers was used to classify the documents and the Naive Bayes classifier had shown the best classification among them.
AmogoUs/Arabic-text-classification
Arabic text documents classified using SVM, k-nn and Naive bayes classifers.
PythonMPL-2.0