CYBER-ANALYTICS: AN EXAMINATION OF MACHINE LEARNING ALGORITHMS FOR SPAM FILTERING
Issues in Information Systems · 2024
This study fills theoretical and application gaps by developing a hybrid spam filtering model that integrates the robustness of random forest classifier with the complex pattern recognition capabilities of neural networks, and the probabilistic reasoning of naïve bayes for enhanced data security and cyber-analytics.We reiterate the significance of spam filtering in addressing cybersecurity challenges and highlight the strengths and limitations of existing techniques; A case is made for the importance of robust spam filtering systems in combating the evolving threat of spam emails.Of the six prediction methods that were initially evaluated, the Random Forest (RF) classifier emerged as the most effective model, achieving the highest accuracy of 95.87% and the lowest misclassification error rate of 4.13%, with balanced performance in identifying true positives and true negatives.The hybridization of Random Forest, Neural Network and Naïve Bayes algorithms further improved accuracy to (97.22%).