Spam has emerged as a significant issue that is endangering the reliability of existing email net... more Spam has emerged as a significant issue that is endangering the reliability of existing email networks. The email has become an AQ1 essential means of sharing information worldwide for personal or commercial purposes. For this reason, creating an effective spam filter is one AQ2 of our biggest challenges. Considering this demand, we build a dynamic spam filter that can filter the standard message and spam messages more efficiently using the most common Naive Bayesian algorithm. Our pro-AQ3 posed model works mainly by considering the content of messages. We used a supervised machine learning model, which contains primarily two phases: Training and Testing. We build a model based on the Bayesian concept in the training phase. In the testing phase, we test our messages by dividing them into words and sentences and calculating their probability for both spam and non-spam categories. Finally, the highest probability value is our desired result and deployed as a web application. Our suggested spam filtering model achieved 98% accuracy. It worked well on both online and offline systems.
Uploads
Books by Utpol Das