In this research study, a cyber-attack prediction model is proposed to address the limitations of existing methods. The primary problem being addressed is the urgent need for an accurate and efficient model to predict cyber-attacks using machine learning. With the increasing frequency and complexity of cyber-attacks, real-time detection and prediction are crucial for minimizing damage and safeguarding sensitive information. The research methodology follows a systematic framework that combines theoretical understanding with practical evaluation to develop effective cyber-security solutions. Results show that the decision tree and random forest models achieved high accuracy and performed well in various evaluation metrics. The random forest model attained the highest ix | P a g e accuracy score of 99.88% and demonstrated excellent precision, recall, f1 score, and ROC AUC score. The proposed model contributes to the development of effective cybersecurity solutions by enhancing the ability to detect and prevent cyber-attacks using machine learning techniques, ultimately improving cybersecurity maturity. Keywords: Cyber-attack prediction, Machine learning, Real-time detection, Decision tree and Random forest models, Cybersecurity solutions