Desenvolvimento de um Sistema Preditivo para Classificação de Emails como Spam ou Não Spam Utilizando Máquinas de Vetores de Suporte (SVM)

Abstract

With the rapid growth of digital communication, the receipt of unwanted messages known as spam has become a significant problem in the information age. This issue affects various sectors, from individual users to large corporations, which face the need to protect their networks and data against this type of threat. Currently, some solutions have been developed to identify patterns in messages deemed spam; however, these solutions have limitations and do not detect all spam effectively. To address this problem, we propose a predictive system that utilizes Machine Learning (ML) techniques. Our goal is to classify messages with high precision, identifying the likelihood that a message is indeed spam, thereby alerting users about the chances of it being spam. We believe that this can help combat the issues that many people are facing.