• Santhosh K
• Salman J
• Sanjeevkumar N
• Jayasmruthi A

Driving behavior analysis is pivotal for enhancing road safety, optimizing fuel consumption, and understanding vehicle performance. The study focuses on developing a model to classify driving behavior using data collected from various vehicle parameters. The dataset comprises driving data from multiple drivers, encompassing features such as fuel consumption, engine speed, vehicle speed, and gear usage. The project is segmented into several phases. Initially, the dataset undergoes preprocessing, followed by the application of feature extraction techniques to derive meaningful features associated with driving behavior. Subsequently, the dataset is divided into training and testing sets, with the majority of drivers' data used for model training and a subset reserved for model evaluation. Various machine learning algorithms are evaluated for model development, including Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Gradient Boosting, Perceptron, and Stochastic Gradient Descent (SGD) classifiers. The performance of each model is assessed using accuracy metrics and classification reports. Additionally, oversampling techniques are employed to address class imbalance issues in the training dataset, ensuring robust model performance across different driving behaviors. The developed models demonstrate high accuracy in classifying driving behavior, with Gradient Boosting emerging as the top-performing algorithm. The model achieves accuracy scores ranging from 81.8% to 99.4% across different classifiers, showcasing its effectiveness in predicting driving behavior. Furthermore, the trained models are tested using data from new drivers, confirming their ability to generalize to unseen driving behavior patterns. The predictions provide valuable insights into the driving styles of the new drivers, facilitating personalized feedback and interventions for improving road safety and fuel efficiency. Overall, the study highlights the potential of machine learning techniques in analyzing driving behavior and underscores their significance in promoting safer and more efficient driving practices.

OBD, Behavior, Accuracy, Google Colab, Grading Boost

"Driving Behavior Classification by Vehicle OBD data Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 3, page no.e834-e840, March-2024, Available :http://www.jetir.org/papers/JETIR2403500.pdf

"Driving Behavior Classification by Vehicle OBD data Using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 3, page no. ppe834-e840, March-2024, Available at : http://www.jetir.org/papers/JETIR2403500.pdf