Spam SMS Classifier Using Machine Learning Algorithms

Spam via short messaging service (SMS) is becoming more prevalent as spammers use it as a means of reaching their targets. While spam SMS messages can be annoying, they can also be dangerous since they might attempt to steal personal information or contain links to malicious websites. The status of SMS spam detection is surveyed in this research. We go over the various kinds of SMS spam; characteristics that help identify spam, and machine learning techniques that have been employed to address the issue. We also note some of the difficulties and potential paths forward in the identification of SMS spam. Support Vector Machine (SVM): SVMs are a class of machine learning algorithms that are applicable to applications involving both regression and classification. SVMs function by locating a hyper plane in the data that has the greatest margin of separation between the various classes. Multinomial Naive Bayes (MNB): A kind of Naive Bayes classifier called MNB is made especially for classification problems involving categorical data. If the data's features are independent of one another, MNB applies the Bayes theorem to determine the likelihood of each class given the data's features. Extra Trees Classifier (ETC): The random forest method serves as the foundation for the ensemble learning algorithm known as ETC. To arrive at a final prediction, ETC builds several decision trees using arbitrary subsets of the data and averages their forecasts. Random Forest Classifier (RFC): Based on the decision tree method, RFC is another kind of ensemble learning algorithm. In order to get at a final prediction, RFC builds several decision trees using arbitrary subsets of the data and averagetheirforecasts. To enhance the overall performance of the model, the predictions made by each individual classifier are combined using the Voting Classifier and Stacking Classifier ensemble models.