A proposed data science approach for email spam classification using machine learning techniques

With the facility of email being accessible to any individual with an internet connection, the proliferation of spam emails is one of the biggest problems which plagues our globally integrated communication systems. The various solutions to filter and hide spam previously included the manual detection of specific keywords and the blacklisting of certain domains created to send spam. However, these methods have certain shortcomings in classifying whether emails are spam or ham. This proposed system attempts to use machine learning techniques to detect a pattern of repetitive keywords which are classified as spam. The system also proposes the classification of emails based on other various parameters contained in their structure such as Cc/Bcc, domain and header. Each parameter would be considered as a feature when applying it to the machine learning algorithm. The machine learning model will be a pre-trained model with a feedback mechanism to distinguish between a proper output and an ambiguous output. This method provides an alternative architecture by which a spam filter can be implemented.