A Novel Compound Normalization‐Based Feature Transformation Learning for Driver Identification

ABSTRACT

Personalized driving assistance, security, and usage‐based insurance models all depend on driver identification in intelligent transportation systems. Primarily, the conventional techniques for driver identification rely on biometric or manually engineered feature data. This process may not accurately capture the subtleties of a unique driving pattern. The aim of this study is to develop a hybrid model, Novel Compound Normalization (NCN), which incorporates the weighted mathematical transformations of three significant normalization techniques, including min‐max, decimal, and . Furthermore, the study compares its performance with the baseline normalization method and explores the statistical significance of the proposed model through hyperparameter tuning of normalization weights. The proposed model integrates three basic normalization techniques through a weighted combination process. The normalized features are subsequently aggregated, and the corresponding weights are determined in one of two ways: If no specific preference exists, use equal weights . Otherwise, adjust the weights based on the reliability or relevance of each dataset. We achieved an accuracy of 98.00%, a precision of 99.00%, a recall of 99.00%, and an F1‐score of 99.00%, outperforming baseline techniques like Min–Max (93.76% accuracy), Decimal Scaling (93.73%), (89.72%), and Z‐score (68.00%) by up to +4.24%. Cross‐validation further confirmed its robustness, with mean accuracy and Macro‐F1 improvements of +2.92% over baselines, supported by statistical significance via Wilcoxon signed‐rank tests () and large effect sizes (Cohen's ). Ablation studies validated the contribution of each component, showing performance drops when any single normalization was removed. The proposed NCN method ensures a scalable, nonintrusive approach for intelligent transportation systems. This emerging technique performs more efficiently in driver identification and autonomous vehicles.