Post-hoc analysis of relevant spatiotemporal features for speech decoding using a linear support vector classifier and LDA. The post-hoc analysis is made by using the SHAP technique
linear-models fnirs interpretable-machine-learning shap shapley-value relevant-feature-analysis neural-decoding shapley-additive-explanations speech-decoding perceived-speech
-
Updated
Jun 10, 2025 - Python