Spam Mail Detection Using Relevance Feature Discovery

Electronic mails are used very widely to share the information quickly. Various domains like business, organizations, academics, political and social uses Electronic mails as they are playing a crucial role in daily communications. The mails which cause the insecurity to the data are called as Spam mails. Spam mails are rapidly spreading all over the mail systems. They may lead to the financial loss and cause the inconvenience to the recipients. To handle these issues the spam mails should be detected efficiently and an appropriate action should be taken on them. Sometimes the spam filters are not able to capture the spam mails accurately as these filters capture data from only particular part of the mail. Relevance Feature discovery is an effective and latest approach for pattern mining. This carries out the mining of the positive, negative and general patterns. This approach includes the various processes like Text processing, Sequential pattern detection, assigning weights to patterns and then saving it to data sets. To filter the emails which are the spam emails efficiently a new approach is proposed. It is based on an innovative Relevance Feature Discovery model. It will scan through the contents of all mails and it will separate patterns in to positive, negative or general categories. Then it will analyse whether they are spam mails or not depending on the type of patterns and process accordingly. It will also synchronize with the Email server and manage emails for users on their system. It will detect the attached image to segregate it into spam or ham image.