- Mail classifier made using flask that classifies unread mails from authorized Gmail account by IMAP4 protocol.
- Training datasets contains only two columns: subject & category and having 20 different categories like junkfile, online trading, management, corporate, etc.
- Used SGDClassifier for trainning the dataset.
- Model Accuracy: 90.6%
- Used Tfidf Transformer & CountVectorizer for feature extraction in ML.
- Used Gridsearchcsv for model selection.
ngandhi369 / ai-email-classifier Goto Github PK
View Code? Open in Web Editor NEWFlask web app made using machine learning model. It uses mails from authorized user's Gmail and shows mails with categorical label on web app based on the mail messages using preprocessed machine learning model on training dataset.