Adaptive intelligent learning approach based on visual anti-spam email model for multi-natural language

Abstract Spam electronic mails (emails) refer to harmful and unwanted commercial emails sent corporate bodies or individuals cause harm. Even though such are often used for advertising services products, they sometimes contain links malware phishing hosting websites through which private information can be stolen. This study shows how the adaptive intelligent learning approach, based on visual anti-spam model multi-natural language, detect abnormal situations effectively. The application of this approach is spam filtering. With learning, high performance achieved alongside a low false detection rate. There three main phases functions intelligently ascertain if an email legitimate knowledge that has been gathered previously during course training. proposed includes two models identify emails. first type language. New trainable Naive Bayes classifier also proposed. trained types languages (Arabic, English Chinese) language use label next model. second built by using classes (phishing normal each language) as training data. (Naive classifier) applied final decision approach. strategy implemented Java environments JADE agent platform. testing AIA involved dataset made up 2,000 emails, results proved efficiency in accurately detecting filtering wide range our suggest performed ideally when tested database biggest estimate (having general accuracy 98.4%, positive rate 0.08%, negative 2.90%). indicates algorithm will work viably off chance, connected real-world database, more common but not largest.