Abstract

Cybersecurity threats in India have increased significantly over the past decade, affecting multiple industries including IT, banking, government, healthcare, and education. This study analyzes a dataset of 308 cyber incidents collected during the years 2015–2024, focusing on attack types, target industries, financial loss, and number of affected users, attack sources, security vulnerabilities, and defense mechanisms. Machine learning models, including Logistic Regression, Random Forest, and Support Vector Machines, along with deep learning using LSTM networks, were applied to classify and predict cyber-attacks. Visualization techniques, such as heatmap and word clouds, were used to explore patterns in the dataset and to highlight the prevalence of different attack types and security vulnerabilities. The results indicate that ransomware, phishing, SQL injection, and insider attacks are predominant, while vulnerabilities like unpatched software and weak passwords are most frequently exploited. The study provides insights into the effectiveness of various models in predicting cyber threats and underscores the importance of proactive cybersecurity measures across sectors in India.

Authors

R. Arunadevi¹, G. Manimannan², R. Lakshmi Priya³
Vidhya Sagar Women’s College, India¹, St. Joseph’s College (Arts & Science), India², Dr. Ambedkar Government Arts College, India³

Keywords

Cybersecurity, Machine Learning, LSTM, Cyber Threats, India, Visualization