A Methodological Study of SMS Spam Classification Using Machine Learning Algorithms
Nitish Sharma · 2022 2nd International Conference on Intelligent Technologies (CONIT) · 2022
Despite the increasing growth of protocol-predicated messaging systems, SMS remains a debatable communication medium in contemporary societies. For its representation, several firms regard SMS to be preferable to emails. Spammers have used SMS to attract the attention of mobile phone users. By delivering SMS to the link or personally contacting the target, the attacker can steal personal data. Therefore, developing a system is mandatory to detect spam. In the present study, with the help of the machine learning, the accuracy of six different models namely Random Forest, Gradient Boosting, Extra Trees, Logistic Regression, Support Vector Machine, and Multinomial Naive Bayes were evaluated and compared using various pre-processing methods that include lemmatization and stemming, subsequently, TF-IDF extraction method was utilized to conclude better pre-processing method and classifier based on ROC AUC curve, accuracy score, f1 score, and balanced accuracy score.