Abstract:
As the vast increases of the electronic mail (email) usages continue, spam (unsolicited bulk mail) has continued to grow because of it is a very inexpensive method of advertising. These unwanted emails can cause serious problem by filling up the email box and thereby leaving no space for ham (legitimate email) to pass through. Case- based filter can adapt to filter new spam by adding new spam case to the case base. Thus, case-based spam filters are suitable for spam filtering because of the dynamic nature of spam. In this paper, a spam filtering system is implemented by using case-based reasoning approach. K-nearest neighbor algorithm is used as case retrieval and case adaption. Edited nearest neighbor rule is used as case maintenance.
