• Raj Sinha
• Reema Jain

Spam filtering has remained a challenge in electronic mail communication because spammers develop new ways to bypass conventional filters with remarkable regularity. In this regard, the Naive Bayes algorithm has been an important cornerstone of most spam-detecting systems due to its simplicity and effectiveness in text classification. This paper reviews the practical application of the Naive Bayes algorithm in spam filtering, covering both theoretical underpinnings and practical implementations, along with performance compared to other algorithms. We discuss the latest developments based on Naive Bayes: hybrid models, ensemble methods, and more advanced feature selection techniques. In this paper, we tackle issues created by the conditional independence assumption of the algorithm and the dynamic nature of spam. This review covers an in-depth study of available literature and experimental studies as proof of the fact that the Naive Bayes algorithm still has much relevance and promise in modern spam filtering systems. We also discuss future research directions toward an integration of the innovations of machine learning to strive further for improvements in accuracy and efficiency of spam detection.

Naive Bayes, spam filtering, email classification, machine learning, text classification, hybrid models, ensemble methods, feature selection, conditional independence, spam evolution

"Next-Generation Spam Filtering: A Review of Advanced Naive Bayes Techniques for Improved Accuracy", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.4, Issue 10, page no.58-67, October-2017, Available :http://www.jetir.org/papers/JETIR1710166.pdf

"Next-Generation Spam Filtering: A Review of Advanced Naive Bayes Techniques for Improved Accuracy", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.4, Issue 10, page no. pp58-67, October-2017, Available at : http://www.jetir.org/papers/JETIR1710166.pdf