Detecting and Mitigating Insider Threat Attacks in Cloud using Machine Learning

The proliferation of smart devices has led to an exponential increase in cyber threats, posing significant challenges to cybersecurity, particularly with the centralized nature of cloud computing. Insider threats, leveraging their privileged access, exacerbate these risks, potentially resulting in valuable data breaches. To address this, a novel approach leveraging machine learning is proposed to detect and classify insider threats, aiming to enhance accuracy through ensemble learning techniques. The study utilizes a customized dataset sourced from multiple files of the CERT dataset, reflecting real-world scenarios.Evaluation of Machine learning algorithms, including Decision Tree classifier, Naive Bayes, KNN Classifier, Logistic Regression, and SVM are evaluated on this dataset. Among these, Support Vector Machine (SVM) emerges as the top performer, achieving an impressive accuracy rate of $98 \%$. This outcome underscores the efficacy of SVM in identifying and classifying insider threats, highlighting its potential as a key component in cybersecurity frameworks. Implementing such approaches can enhance organizations' resilience against insider threats, safeguarding sensitive data and mitigating potential risks to business operations.