Machine Learning in Cybersecurity: Evaluating Text Encoding Techniques for Optimized SMS Spam Detection

SMS spam poses serious online security threats, including phishing and malware risks. Effective detection and prevention are vital for user protection. This study aims to improve SMS spam detection accuracy by exploring efficient text encoding and classification methods. We assess three text encoding techniques: Bag of Words (BoW), Term Frequency-Inverse Document Frequency (TF-IDF), and Word2Vec. We conduct exploratory analysis, transform messages into numerical vectors, and enhance classification with additional features. Comparative analysis using various metrics helps select the best-performing algorithm for model construction. The research has practical implications, boosting spam prevention and enhancing cybersecurity in text messaging.