Spam Detection With Naïve Bayes And TF-IDF

This paper, titled "Spam Detection With Naïve Bayes And TF-IDF," explores an approach to detecting spam emails by applying Natural Language Processing (NLP) techniques combined with machine learning methods. The study used a balanced dataset of 2,000 emails (1,000 spam and 1,000 non-spam) collected from Kaggle. After text preprocessing, feature extraction was performed using the Term Frequency–Inverse Document Frequency (TF-IDF) method. The Naïve Bayes algorithm achieved 98.86% accuracy and 98.35% F1 score, demonstrating its effectiveness in spam email classification. This article was originally published in the IOSR Journal of Computer Engineering (IOSR-JCE), Volume 27, Issue 4, Series 2 (July–August 2025), pp. 47–51. Original publication link: https://www.iosrjournals.org DOI: 10.9790/0661-2704024751