Evaluating Cyber-Crime Using Machine Learning and AI Approach for Environmental Sustainability

Evaluating Cyber-Crime Using Machine Learning and AI Approach for Environmental Sustainability

연구 분야: Cryptography

학회: International Conference on Sustainable Development through Machine Learning, AI and IoT

The online world has become an essential part of everyday life in India due to its fast expansion and technical improvements. But as technology has advanced, cyber-crimes have also increased dramatically, posing serious risks to the economy, security, and privacy. This research examines the intricate terrain of cybercrimes targeting minors in India, scrutinizing more than 16 lakh cases that have been documented during the three years preceding this one. The study takes an extensive approach, applying machine learning modelling such as Gradient Boosting, AdaBoost, Random Forest, Gaussian Naive Bayes, Bernoulli Naive Bayes, Linear SVM, Logistic Regression, and Gradient Boosting in conjunction with exploratory assessment of data. The findings show different accuracy rates; Random Forest, Gradient Boosting, Linear SVM, and Gaussian Naive Bayes all show an impressive 88% accuracy. AdaBoost and Logistic Regression reach 75%, but Bernoulli Naive Bayes comes in lower at 62%. The comprehension of computational interconnections is improved with correlation matrices. This thorough analysis provides important insights for academics, policymakers, and stakeholders in the field of child cybersecurity by exposing common cyber threats targeting minors and highlighting the role that algorithms for learning may play in detecting and averting these risks for environmental sustainability.