From VoIP vulnerabilities to AI defenses: A survey on fraud call prevention
Agrim Ray, Ashutosh K. Singh · 2025
Spam calls, or scam calls, are serious threats as they trick individuals and divert emergency services away from life-threatening situations. Increased Voice over Internet Protocol communication has necessitated fraud detection as a means of safeguarding against financial loss, identity theft, and service abuse. This study investigates artificial intelligence-based methods for fraudulent call detection using Natural Language Processing and machine learning methods to examine patterns and detect abusive patterns. In this paper, an attempt has been made to highlight various researchers that are done to analyze how machine learning, deep learning and artificial intelligence has been used to detect spam calls or spit calls. This study explores how VoIP networks have been analyzed to detect such fraudulent schemes based on various call behaviors. Based on call metadata, speech patterns, frequency analysis, and anomaly detection, researchers have developed AI-driven techniques to detect and counter VoIP-based attacks.