Adaptive privacy policy prediction for email spam filtering

Internet being an expansive network of computers is unprotected against malicious attacks. Email that travels along this unprotected Internet is eternally exposed to electronic dangers. Businesses are increasingly relying on electronic mail to correspond with clients and colleagues. As more sensitive information is transferred online, the need for email privacy becomes more pressing. Spam mails eat up huge amounts of bandwidths and annoy the receivers. Unsolicited messages are often used to compel the users to reveal their personal information. Spam mails are commonly used to ask for information that can be used by the attackers. Email is a private medium of communication, and the inherent privacy constraints form a major obstacle in developing efficient spam filtering methods which require access to a large amount of email data belonging to multiple users. To alleviate this problem, we foresee a privacy preserving spam filtering system that is adaptive in nature and help the user to compose privacy settings for their emails. We propose a two level framework which filters spam and also determines the best available privacy policy. Spam detection is done by similarity matching scheme using HTML content and the adaptive privacy framework enables the automatic settings for email that are filtered as spam.