what is the main challenge associated with the curse of dimensionality in clustering? a$)$ decreasing risk of overfitting b$)$ enhanced interpretability c$)$ increased computational complexity d$)$ improved cluster separation