Leveraging artifact reputation analysis and contextual sentiment analysis for advanced detection of vishing and smishing attacks

Abstract

The rise of advanced mobile technology has brought about the widespread presence of mobile devices in our society. These portable and versatile gadgets have become essential items for individuals due to their convenience and capabilities. As technology continues to play a pivotal role in modern life, an evergrowing number of people rely on mobile devices for almost all life activities including crucial financial activities and business routines. However, the increasing popularity of mobile devices has also exposed users to a heightened risk of falling victim to fraudulent schemes. Perpetrators have been exploiting mobile users by pretending to present authentic and legitimate requests and opportunities, leading to the divulgence of personal and sensitive information. These deceitful activities have seen a significant increase, affecting individuals of various ages, educational backgrounds, and levels of technological literacy. Additionally, malicious actors employ advanced methods to conceal their identities, making it challenging to prevent and counter these attacks. Two prevalent yet under-addressed issues in this context are vishing and smishing. This research study introduces a system designed to detect vishing and smishing attempts more accurately. The system analyzes the reputation of suspicious artifacts in messages and call conversations using third party threat intelligence services. Further, it employs natural language processing and machine learning techniques to examine the content of voice calls and SMS messages. It identifies suspicious elements such as keywords and phrases commonly used in phishing attacks, sensitive information as well as the context of the content.