SHapley Additive exPlanations (SHAP) for Efficient Feature Selection in Rolling Bearing Fault Diagnosis

Abstract

This study introduces an efficient methodology for addressing fault detection, classification, and severity estimation in rolling element bearings. The methodology is structured into three sequential phases, each dedicated to generating distinct machine-learning-based models for the tasks of fault detection, classification, and severity estimation. To enhance the effectiveness of fault diagnosis, information acquired in one phase is leveraged in the subsequent phase. Additionally, in the pursuit of attaining models that are both compact and efficient, an explainable artificial intelligence (XAI) technique is incorporated to meticulously select optimal features for the machine learning (ML) models. The chosen ML technique for the tasks of fault detection, classification, and severity estimation is the support vector machine (SVM). To validate the approach, the widely recognized Case Western Reserve University benchmark is utilized. The results obtained emphasize the efficiency and efficacy of the proposal. Remarkably, even with a highly limited number of features, evaluation metrics consistently indicate an accuracy of over 90% in the majority of cases when employing this approach.

Behavioral Adaptations in Tropical Dairy Cows: Insights into Calving Day Predictions

“…The SHAP summary plot in the present study identified activity, rest time, and rest per bout as the most crucial features for developing an effective calving day prediction model. The SHAP summary plot illustrates each dot as a single SHAP value for a specific feature within a single data instance [ 42 ]. Notably, the clustering of dots reflects the strength of a feature’s interaction with the model’s outcomes [ 43 ].…”

“…The SHAP summary plot in the present study identified Using a Shapley Additive Explanation (SHAP) model, the present study evaluated the importance of features for calving day predictions in ML models. This model leverages concepts from game theory [42] and integrates seamlessly with the random forest regressor model in the present study [43] to illustrate the contribution of each input variable to the prediction of calving day models. Figure 10 presents the mean absolute SHAP values for each feature.…”

Effect of soil characteristics on compressive strength of cement stabilized earth blocks: Statistical analysis and Artificial Neural Networks model

“…These models learn the relationships between different material properties (independent variables) and compressive strength (dependent variable), allowing researchers to identify the most signi cant factors in uencing strength. Also, techniques like SHAP (SHapley Additive exPlanations) values within machine learning models pinpoint which speci c features within the material data contribute most signi cantly to the predicted strength [22].…”