PHISHING ATTACKS IN INDIA: AN ANALYSIS OF EMERGINGTECHNIQUES, DETECTION STRATEGIES AND 2025 FRAUD TRENDS

Abstract

Phishing is a deceptive cybercrime in which attackers impersonate as an authenticateorganizations, person or services to gain any individual’s trust and obtain their private information suchas financial credentials, login details, net banking accounts and other useful information. Thiscybercrime has moved beyond conventional fraudulent emails and now operates across SMS, voicecalls, social-media platforms, fake websites, QR codes and other digital communications. I reviewedearlier studies on phishing attacks and detection, synthesizes their main findings, and places themagainst the Indian cyber-fraud environment reported by government agencies in 2024-25. The reviewshows a clear progression from rule-based and blacklist-oriented protection towards machine-learning,natural-language processing, pattern analysis and hybrid detection. Random Forest and Support VectorMachine approaches repeatedly show strong performance in the reviewed literature, while NLP-basedanalysis is useful for extracting linguistic indicators from messages. As phishing data and attackerapproach change over time so highly precise data is also vulnerable. Indian government recordsunderline the scale of the wider cyber-fraud problem: the Ministry of Home Affairs reported 3,822,550cyber-fraud incidents through the National Cyber Crime Reporting Portal from inception to January2025, with 3,644,819.85 lakh reported as the amount involved. In a separate 2025 government replyreported that the Citizen Financial Cyber Fraud Reporting and Management System had helped savemore than 5,489 crore across more than 17.82 lakh complaints. CERT-In also reported a majornational cyber-incident workload for 2024-25. The paper argues for a layered model combining senderauthentication, URL and content analysis, machine learning, signals, threat intelligence, user awarenessand rapid reporting. Rather than treating phishing as only a technical classification problem, the studypresents it as a socio-technical problem in which technology and human judgement must work togetherto achieve so concrete solution. Keywords: Phishing; cyber fraud; social engineering; smishing;vishing; machine learning; NLP; Random Forest.