Spam Detection Using Machine Learning: A Logistic Regression Approach

Spam emails pose a significant challenge to digital communication, leading to security threats and reduced productivity. This paper proposes a machine learning-based strategy to spam detection using Logistic Regression with TF-IDF vectorization. The dataset is prepared by handling missing values and normalizing labels. A TF-IDF model with bigram inclusion is implemented for feature extraction, followed by a balanced Logistic Regression classifier to address class imbalance. Experimental results indicate promising accuracy, demonstrating the effectiveness of the proposed method. Future enhancements include deep learning models and ensemble techniques for improved spam classification.