svm-spam-filter
A python implementation of a simple spam classifier using support-vector machines. Requires python3.6+ to run.
Usage
First obtain enough emails to train the classifier. For example, use the SpamAssassin public mail corpus: https://spamassassin.apache.org/old/publiccorpus/. Place spam in a folder called spam and non-spam in a folder called ham. Then, train the classifier:
python3 main.py
Note that preprocessing of the emails might take a while, depending on the amount of the emails that are used to train the classifier. To classify your own emails (after you have trained the classifier), run the following:
python3 predict.py FILE
Where FILE contains the source code of the email you want to classify.