Explainable AI for Hand Gesture Recognition

Understanding why models make decisions.

Most machine learning systems focus exclusively on accuracy. This project explored a different question: can we understand why a model predicts a particular hand gesture?

Using surface electromyography signals collected from hand movements, multiple machine learning models were trained and compared. LIME-based explanations were then used to identify which signal characteristics influenced each model's predictions.

The result was an interpretable machine learning framework that balanced predictive performance with transparency. This work was later published at an IEEE conference.

Python · Scikit-Learn · LIME · Random Forest · Gradient Boosting · Explainable AI