Spam / Junk E-Mail Filter Technique

Most e-mail readers spend a significant amount of time regularly deleting junk e-mail (spam) messages, which are a part of marketing campaigning efforts of various companies wherein users normally signed in and it also results in increasing volume of storage space and consumes network bandwidth. A challenge, therefore, rests with the developers and improvement of automatic classifiers that can differentiate authentic e-mail from spam. Spam detectors normally use Naïve Bayesian approach and large feature sets of binary attributes that determine the existence of common keywords in spam emails. Spammers/Marketers recognize these approaches to impede their messages and have developed tactics to bypass these filters, but these ambiguous tactics are themselves patterns that human readers can often identify quickly. The preliminary study tests an alternative approach using a neural network (NN)