The whole business of sending electronic messaging spam started with Gary Thuerk, a marketeer for the Digital Equipment Corporation (DEC). On May the 3rd, 1978 he sent an email to about 600 recipients on the ARPANET network (an early version of the Internet) selling information about the computers available at DEC. It didn’t go down well, though a few people were interested in the products he was selling. Today approximately 50% of email messages are spam, and the problem is also significant in SMS, Whatsapp and social media platforms such as Twitter and Facebook. As well as being an annoyance, spam messaging can be used to spread malware and link to fraudulent sites in order to steal passwords and credit card information. In this article we are going to demonstrate how a spam filter can be developed using some relatively simple machine learning techniques. The whole implementation takes 9 lines of code using Python, Pandas and scikit-learn.
See my full post at Open Data Science News
Subscribe to SimplyML: Simply Machine Learning
Get the latest posts delivered right to your inbox