MOTORBIKE ACCIDENT SEVERITY PREDICTION USING MACHINE LEARNING ALGORITHMS

MOTORBIKE ACCIDENT SEVERITY PREDICTION USING MACHINE LEARNING ALGORITHMS

ABSTRACT

Of all traffic accidents, motorcycle accidents are the most dangerous. Bicycle accidents cause a lot of fatalities. Pedestrians and private property are both harmed in motorcycle accidents. The relatives of the victim must pay a significant sum of money for the biker’s medical care. To secure funding for the victim’s therapy, their family may occasionally have financial difficulties. Awareness of accident prevention and avoiding factors that lead to accidents are essential for bikers. Various machine learning algorithms have been used in this work to predict the seriousness of motorcycle accidents. Then, we gather information based on certain parameters, including, but not limited to, speeding, overtaking, turning, bike fitness difficulties, unsignaled speed breakers, hazardous lane changes, conversing with a passenger, and highways without road barriers. Our only assemble data from motorcyclists who have experienced mishaps. Once the data had been gathered, we processed it all and created a processed dataset. On the cleaned-up dataset, we applied machine learning methods. Since the beginning of prediction and detection systems, machine learning has been used in many of these systems. We use a variety of methods, including Random Forest, Multilayer Perception (MLP), Decision Tree, Logistic Regression, k-Nearest Neighbors (KNN), AdaBoost, GNB, SVM with RBF Kernel, Linear SVC, and Gradient Boosting. The most accurate results were provided by MLP. An 83.10 percent accuracy was displayed. In terms of sensitivity, specificity, F1-Score, and precision, MLP once again outperforms other methods.

MOTORBIKE ACCIDENT SEVERITY PREDICTION USING MACHINE LEARNING ALGORITHMS. DOWNLOAD SOURCE CODE NOW

Sharing is caring!