EVALUATION OF MACHINE LEARNING ALGORITHMS FOR FILTERING AND ISOLATING SPAMMED MESSAGES
No Thumbnail Available
Azeez, Nureni Ayofe
FUTA JOURNAL OF RESEARCH IN SCIENCE
The use of mobile application is rising on daily basis as they offer a wide range of services and at the same time elevating the fee of those services. Short Message Service (SMS) is considered one of the most commonly deployed messaging systems. However, this deployment has led to a spike in attacks on mobile devices such as SMS Spam. In this article, Artificial Intelligence (AI) method, which discovers and filters unsolicited spam messages was adopted. The machine learning algorithms used are Logistic Regression, Decision Trees, Gaussian Naïve Bayes; Multilayer perceptron (MLP) and Support Vector Machine (SVM). After experimentation, it was observed that MLP Classifier produced the best results with 93.1% true positive rate.
Spammed message; mobile application; artificial intelligence; attacks; SMS