Variable Generalization Evaluation of Supervised Learning Models for Detection of Spam Messages

In the last decade, the use of mobile phones has increased dramatically, leading to new promotions and business advertisements (Spams SMS) being sent to mobile phones. People, unfortunately, give out their mobile phone numbers when using daily services, which fill up spam ads, increasing the demand for data processing. Machine learning is one of the popular domains in computer science that is used to process and classify data. Various algorithms for classifying data have been proposed in the literature. KNN (k-nearest neighbors), SVM, and neural networks are just some of the popular algorithms that have been used recently. The purpose of this study is to evaluate and compare the performance of different classification algorithms based on accuracy and F1 Score. For this study, seven classifiers are applied to three different real data sets and corresponding results are presented.