Browsing by Author "Sanjay, Kumar N"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    SPAM Filtering Using the Social Anthropology and Data Mining Technique
    (2015-12-10T07:04:56Z) Sanjay, Kumar N
    The rapid growth of the spam emails has incited the development of many spam’s filtering mechanism. The traditional Bayesian spam filter is very stable to static keywords. It can continuously evolve to embark upon new spam by learning keywords in new spam emails. Nevertheless, Bayesian spam filters can be easily deceived by using synonyms for the spam keywords in the emails. Moreover they depend on user feedback and require significant amount of time to become acclimatize to a new spam keyword used in spam emails. SOcial Aided Personalized (SOAP) and effective spam filter primarily focused on the social relationship of the email correspondents. It used a SOAP agent to collect all the user information and worked in a distributed manner to fight against the spam emails. Although SOAP improved the performance of Bayesian spam filter greatly, time complexity was one of the major drawbacks. In addition to this, it was a client side solution rather than the centralized solution leaving service providers clueless about spammers. In order to develop an accurate and user-friendly spam filter, here we propose Social Anthropology and Data Mining (SADM) spam filter. Unlike the previous spam filter SADM is not restricted to the client side or focus only on parsing keywords, rather is a server side solution accounting the social relationship of the email correspondents. SADM has four major processing steps: Members list, Member’s friend list, Global list and Content-based data mining. SADM completely eliminates the false positive rate and filter the spam efficiently.