Fig. 2 | Scientific Reports

Fig. 2

From: Identifying and predicting gait stability metrics in people with stroke in uneven-surface walking using machine learning

Fig. 2

SHAP Value Plot. The x-axis represents the SHAP value of each feature, indicating its contribution to the predicted class. In this visualization, SHAP values are displayed separately for HC (left) and PwS (right), where positive values indicate movement of the model output toward correctly predicting the displayed class (e.g., positive SHAP values on the HC panel increase the likelihood of being classified as HC). The color scale indicates feature magnitude (red = high, blue = low). For example, a high-value feature (red) located on the positive side of the HC plot indicates high feature values increase classification toward HC, whereas a similar pattern in the PwS plot reflects increased likelihood of PwS classification. (a) RandomForest (GAN(N = 1000)); (b) Support Vector Classifier (ctGAN(N = 1000)).

Back to article page