SMART FILTERS FOR SMS SPAM: A MACHINE LEARNING APPROACH TO SMS CLASSIFICATION

Abstract

The exponential rise in the number of undesired text messages delivered via SMS has been directly related to the explosion in the number of mobile phones sold. Although various information channels are considered "spotless" and trustworthy in many parts of the world, ongoing reports show that cell phone spam is significantly increasing. It is a big problem. It is becoming increasingly pervasive worldwide, especially in Asia and the Middle East. In the same way that finding a solution to such an issue can be time-consuming, so can the process of identifying spam texts from genuine communications. It solves many difficulties and makes life much easier because it can distinguish between real SMS and spam. In any event, it faces specific challenges and obstacles that are unique to itself. During this current research, we have investigated five Machine Learning (ML) methods to identify spam in a short text message using a single dataset containing SMS spam Collection. The SMS spam dataset was extracted from the Kaggle repository. The experiment is carried out on the R platform. Eleven characteristics, including binary and numeric features like Char Count, Has number, Has URL, Has Date, Has dollar, Emoticon, Email, and Phone, as well as spam count, ham count, and spam binary, are employed in this research. These features are used for feature selection and showing results using Machine Learning(ML) approaches. The effectiveness of the various strategies or methods is evaluated using metrics such as sensitivity, accuracy, precision, F1 score, recall, and specificity. The outcomes show that the light gradient boosting machine (LGBM) with these features achieved a sensitivity score of 100, precision score of 100, F1 score of 100, recall of 100, and specificity score of 100, with an optimal accuracy score of 100 percent, which is outstanding compared to all other state-of-the-art studies.