Email Spam Classification Using Novel Fusion of Machine Learning and Feed Forward Neural Network Approaches
Keshetti Sreekala, Maganti Venkatesh, M. V. Ramana Murthy, S. Venkata Meena, Srinivas Rathula, Annemneedi Lakshmanarao · 2025
In the age of information, electronic conversation has become an integral unit of our daily lives. However, this satisfaction is accompanied by the continual challenge of email spam, which not only inundates inboxes but also presents a severe peril to cybersecurity. Email spam rests a pervasive issue, with usual spam filters regularly struggling to adapt to evolve spamming techniques. This paper intends to present progressive advanced ML and DL techniques. This paper introduces an innovative solution that leverages a strategic ensemble of the two Machine Learning (ML) and Feed Forward Neural Network (FNN) to improve the accuracy and efficiency of email spam categorization. By employing cutting-edge algorithms and models, the goal is to build a robust and resilient system capable of strongly identifying and filtering out spam emails. A dataset from Kaggle with “spam” and “no spam” are collected. Later, various ML classifiers namely SVC, KNN, NB, DT, RF, FFNN and LGR are applied and achieved better results. Later, fusion of Machine Learning best classifier random forest and Deep Learning FFNN applied with both types of ensembles namely stacking and voting classifiers and achieved accuracies of 98.9% and 99%.