Machine Learning Techniques on Mobile SMS Spam Detection
Megha Birthare, Neelesh Kumar Jain, Alpana Meena · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2025
Unsolicited mass sms or fraudulent sms delivered to people or organisations are known as spam.To prevent data breaches and invasions of privacy, spam texts must be recognized and eliminated.Scholars are consistently investigating machine learning approaches and strategies to efficiently distinguish and categorise spam sms from authentic ones, often known as "ham" sms.Researchers have built systems that can accurately classify sms as spam or ham by analysing numerous textual elements.This study assesses the accuracy of several classification techniques in identifying spam from valid sms by analysing data gathered from multiple sources.sms are filtered and categorised using Natural Language Processing (NLP) algorithms according to their content.The Extreme Learning Machine (ELM) is one instance of a machine learning model used for this purpose.ELM is the state-of-the-art feedforward neural network technique with a single hidden layer.ELM avoids overfitting problems and has quick training times compared to standard neural networks.Because ELM only needs one iteration cycle, spam detection using it is both practical and efficient.This paper concludes by reviewing and contrasting a number of machine learning techniques for spam detection, emphasising the efficiency and adaptability of strategies like ELM in protecting against spam sms on a variety of domains.