This research aims to develop a predictive model for catalytic performance using Artificial Neural Networks (ANN). The catalytic activity is highly depended on various factors such as acidic sites, porosity, surface area etc. during any reaction. For our model, we have selected a unique data set which was used for oxidation of cyclohexane. Here oxidation of Cyclohexane to KA oil was performed in presence of hollow zeolites (Au@ZSM-5, Au@MCM-22, ZSM-5 and MCM-22). Acidic sites present in hollow zeolites were found to be responsible for higher conversion for oxidation of cyclohexane to KA oil. Using regularization techniques and early stopping, we developed a model which achieved a high accuracy (>95%) for predicting unseen test data for oxidation of Cyclohexane to KA oil in presence of hollow zeolites. The results suggest that ANN can effectively model the complex, non-linear relationship between different aspects such as dependence of conversion rate on catalyst acidic sites, providing insights into catalyst behavior.