Mail-Spam-Filtering

It uses machine learning models to predict whether the email is spam or ligitimate.

Best thing would be to follow my blog-post for implementation. The description about the steps to build a spam filter from scratch can be read from my blog:

https://appliedmachinelearning.wordpress.com/2017/01/23/nlp-blog-post/

It is a python implementation using Naive Bayes Classifier and Support Vector Machines from Scikit-learn ML library.

The results has been shown on two publicly open corpus.

1. Ling-spam corpus
2. Euron-spam corpus

The link for corpus/dataset download is given in blog-post.

Note : Directory path used for training and testing models in lingspam_filter.py and euron-spamfilter.py needs to be properly set accordingly.