Exploring the Potency of Machine Learning Approaches in Enhancing Spam Detection Accuracy

Abstract Spam in mobile communication networks is a serious problem since it interferes with user experience and jeopardises user privacy. A precise and reliable spam detection technique is required for mobile Short Message Service (SMS) communication to successfully combat this issue. In our study, we suggest utilising machine learning-based spam detection strategies to achieve precise identification. With this technique, ham and spam messages in mobile communication are categorised using machine learning classifiers including Support Vector Machine (SVM), K Nearest Neighbour (KNN), Logistic Regression (LR), and Light GBM (LGBM). The SMS spam collection dataset, which was split into two groups for training and testing, was used in our trials to assess the methodology. In terms of classification performance, the trials' findings demonstrated that LGBM performed better than others, obtaining a remarkable accuracy of 98%. This shows how successfully we have addressed the problem of spam in mobile SMS communication.