Neutrosophic Ensemble Classification for Uncertainty-Aware Bearing Fault Detection: Evidence from Laboratory and Variable-Speed Industrial Benchmarks

A neutrosophic ensemble combining Random Forest, XGBoost, and Logistic Regression decomposes predictions into four indicators for uncertainty-aware bearing fault detection, evaluated on CWRU and JNU benchmarks under a leave-one-condition-out protocol. On CWRU, the ensemble reaches 100.00 percent accuracy on three of four held-out loads (92.27 percent on the fourth), while on JNU, accuracy drops to 40.64 percent, below a majority-class baseline, with Logistic Regression (57.91 percent) generalizing better than tree ensembles.