Implementing an Agent-based Multi-Natural Language Anti-Spam Model

The spam is a negative practice of illegitimate use to the email services through unsolicited email such as phishing for scam practices which affects the email reliability. Spam problems and its influence on the society have been investigated and discussed from different perspectives. Several studies have looked into the influence of the spam on the economy, financial, marketing, business and management, while others deliberate the impact of the spam on the security and privacy. Subsequently, there are different anti-spam techniques that have spam filtering or blocking mechanisms. This work attempts to investigate an available anti-spam technology and highlight the possible improvements. Consequently, it constructs a new agent-based anti-spam model that can overcome some existing limitations. The Multi-Natural Language Anti-Spam (MNLAS) model comprises visual information, and texts of an email in the spam filtering process. The MNLAS is implemented in a Java environment using Jade agent platform. The application detects and filters spam emails of different types using a dataset of 200 emails.