Mehmet Serdar Güzel

Electric motors are critical components of modern industrial and transportation systems, making accurate fault diagnosis and predictive maintenance essential for operational reliability. This study proposes a unified machine learning framework for motor fault classification, temperature prediction, and remaining useful life (RUL) estimation using three publicly available electric motor datasets. The proposed pipeline integrates sliding-window segmentation, multi-domain feature extraction, consensus-based feature selection, and physically motivated descriptors derived from electrical and thermal measurements. Five machine learning algorithms, namely Random Forest, Support Vector Machine (SVM), XGBoost, LightGBM, and Long Short-Term Memory (LSTM), were systematically evaluated. For fault diagnosis on the Ottawa UOEMD-VAFCVS dataset, evaluated with a file-level train–test split in which all segments of a recording are kept in the same partition, LightGBM achieved the highest performance on the held-out recordings (Accuracy = 1.0000, F1-score = 1.0000, ROC-AUC = 1.0000). Sliding-window segmentation increased the number of trainable observations from 128 to 2,432 and improved the macro F1-score from 0.569 to above 0.994 across all models. For thermal prediction on the PMSM dataset, LightGBM obtained R² values of 0.993, 0.991, 0.989, and 0.987 for rotor, stator winding, stator tooth, and stator yoke temperatures, respectively, with mean absolute errors below 1.4°C. In an exploratory experiment on the synthetically generated EVIoT dataset, which contains no genuine run-to-failure trajectories, XGBoost and LightGBM reached R² = 0.933 on a 30-step smoothed RUL target; this value largely reflects the smoothing of the target and is reported only as a demonstration of the pipeline, not as validated prognostic capability. A Friedman test over N repeated runs indicated significant differences among the five classifiers (χ² = 11.556, p = 0.021, Kendall’s W = W). Within the evaluated configurations, feature-engineered gradient-boosting models matched or exceeded the LSTM baseline at low computational cost; a systematic comparison with end-to-end deep architecture is left for future work.

Credit-risk prediction must balance the detection of adverse outcomes with the need to inspect how model predictions are formed. This study presents an AutoML-driven hybrid framework combining a supervised xDNN-inspired representation, traceable class-conditional prototypes, a cost-sensitive Random Forest, and out-of-fold stacking. The framework was evaluated on the Taiwan Default of Credit Card Clients and Give Me Some Credit datasets using three randomly seeded train–validation–test partitions per dataset. Logistic Regression, Random Forest, XGBoost, LightGBM, CatBoost, a multilayer perceptron, and FT-Transformer served as comparative models. Across the three data splits, the proposed hybrid achieved a mean ROC–AUC of 0.7839, a mean average precision of 0.5549, and a mean Brier score of 0.1623 on Taiwan. On Give Me Some Credit, the corresponding mean values were 0.8625, 0.4032, and 0.0491, respectively. At validation-selected F2 thresholds, mean recall was 86.81% and 63.39%, respectively, with corresponding mean F2-scores of 0.6366 and 0.5168. Performance remained close to strong baselines, with substantial false-positive burdens at the selected operating points. TreeSHAP described how the final forest used original and xDNN-derived inputs, while stored prototypes supplied traceable development-data examples. The framework integrates sensitivity-oriented prediction and complementary feature- and example-based evidence within a leakage-controlled workflow.