Mobile Ad-hoc Networks (MANETs) are inherently vulnerable due to its dynamic, decentralized structure and requirement for advanced solutions for security risk mitigation. It is suggested to use a machine learning (ML)-based strategy for effective attack detection and mitigation. Important characteristics are identified and standardized from the network traffic data, such as the time it takes to respond, how frequently replies occur, and the rates at which packets are dropped. The efficacy of six Machine Learning classifiers—Support Vector, Naïve Bayes, K-Nearest Neighbour, Logistic Regression, Machine, Multilayer Perceptron, and Extreme Gradient Boosting—in identifying black hole attacks is assessed through training. The mentioned outcomes exhibit the efficiency and efficacy of the recommended machine learning strategy in limiting black hole attack risks in MANNETs. These results are based totally on numerous factors including Packet shipping Ratio (PDR), overall time, accuracy and energy usage.