IMPROVING SPAM DETECTION ACCURACY WITH RANDOM FOREST AND TEXT VECTORIZATION

With the exponential growth of digital communication, email and messaging platforms have become a major vector for spam. Effective spam detection is essential to enhance user experience, reduce resource consumption, and ensure information security. This paper presents a machine learning-based approach for spam detection using the Random Forest classifier. The model was trained and tested on a benchmark dataset, achieving an accuracy of 87.77%. Random Forest, being an ensemble learning technique, aggregates the predictions from multiple decision trees to improve classification performance and reduce overfitting. The proposed system demonstrates reliable classification of spam and ham messages, showing that Random Forest can serve as a robust baseline method for text-based spam detection. Further analysis reveals that the model balances precision and recall effectively, making it suitable for real-time deployment in email filtering systems.