Spam emails are becoming an ever-present issue that demands powerful and adaptable screening solutions as the usage of digital communication rises at an exponential pace. The goal of this study is to increase the effectiveness and accuracy of email classification by using machine learning techniques to spam filtering. The recommendation approach employs a broad range of features gleaned from email text, headers, and sender information. A range of machine learning approaches are employed to train models that can distinguish between authentic and spam emails. Neural Networks, Support Vector Machines, and Naive Bayes are examples of these algorithms. The dataset utilised for trials is large to ensure that the models are robust and can be used to a wide range of circumstances. Because spam is always changing and spammers strategies are constantly improving. This study show that the machine learning-based spam filtering technology is successful, they also show that it can adapt to new spam pattern.