Abstract:

Short Message Service, or SMS, has become obsolete in this day of widely used instant messaging apps. Instead, it is now the speciality of service providers, corporations, and other organisations that use it to target regular users for spam and marketing. The usage of regional language information typed in English is a recent trend in spam communications that makes it more difficult to identify and filter such messages. An expanded version of a typical SMS corpus that includes both spam and non-spam messages has been utilised in this work. It includes labelled text messages in regional languages, such as Bengali or Hindi, that have been entered in English and collected from local mobile users. Using a collection of features and machine learning algorithms that are often employed by academics, the Monte Carlo technique is used for learning and classification in a supervised manner. The outcomes show how various algorithms perform in successfully tackling the given task.

Description:

.