Neuro-fuzzy systems is a popular hybridization in soft computing that abstracts a fuzzy model from given numerical examples using neural learning techniques to formulate accurate predictions on unseen samples. The fuzzy model incorporates the human-like style of fuzzy reasoning through a linguistic model that comprises if-then fuzzy rules and linguistic terms described by membership functions. However, modeling data using neuro-fuzzy systems involves the contradictory requirements of interpretability versus accuracy. Prevailing research that focused on accuracy employed optimization that resulted in membership functions that derailed from human-interpretable linguistic terms. In addition, the modeling of high-dimensional data requires a large number of if-then fuzzy rules that exceeds human level interpretation. This thesis focuses on increasing interpretability without compromising accuracy using a novel hybrid intelligent Rough set-based Neuro-Fuzzy System (RNFS), which synergizes rough set-based knowledge reduction with neuro-fuzzy systems. RNFS directly addresses the problems with the following contributions.