Spam SMS Detection Using Natural Language Processing

A spam SMS is an unsolicited text message dispatched to a large population, illegitimately, without the authorization of the recipients. Reports suggest that the number of Spam SMS received by citizens of any country is approximately 400 million, and this number is increasing day by day. These SMS messages may contain malicious links, causing security threats including phishing attacks and malware infections. Some spam SMS messages can involve fraudulent schemes, leading to financial losses for the receiver and can also cause privacy threats. Hence, there is an urge to detect these spam SMSs correctly and avoid the severe consequences for the recipients. The proposed study emphasizes the development of Natural Language Processing based Machine Learning models to identify spam SMS texts. The process involves tokenization of the SMS texts, stemming of the words, and many more abstractions of the keywords for easier classification. Firstly, the text messages are processed using NLP and then the processed data is trained on different ML algorithms. Multiple ML models are created and tested on various kinds of SMS messages for the detection of spam messages. This expert system ensures accurate and precise detection of received messages. This expert system intends to reduce the threats and safeguard the security and privacy of the consumers, by essentially recognizing and steering clear of spam SMS messages. This system aims to empower users with the tools and knowledge needed to navigate the digital landscape safely and securely.