Analyzing Cybersecurity Incident Trends in India (2020–2024): A Machine Learning Approach for Attack Classification and Financial Loss Prediction

Cybersecurity risks in India have grown significantly within the last ten years because of the pace of digital transformation, the expansion of online services, and the prevalence of the internet. The present paper includes an in-depth examination of cybersecurity incidents in India documented between 2020 and 2024 to gain insights into patterns of attacks, industry, and geographical weaknesses, and financial cost of such attacks through the application of statistical methods and machine learning algorithms. An analysis is conducted on a dataset of more than 1,000 incident records covering various types of attacks, cities, and industries. According to the exploratory data analysis, the number of cyber incidents has been steadily growing, and the dominant attack vectors are ransomware, phishing, and online fraud. The statistical analysis, ANOVA and chi-square test, demonstrate that there is statistically significant variation in financial loss by type of attack and significant correlations of sector and incident type. Moreover, the trained machine learning models, such as Random Forest, Support Vector Machine, and XGBoost, are used to categorize the different types of attacks and estimate the financial losses. The classification accuracy of the proposed models is over 85% and the regression models have R2 values greater than 0.78. The results also offer policy implications to policymakers, cybersecurity practitioners, and organizations to create sector- and regionindustry defense measures.