From last couple of decades, literacy rate is increased all over the globe so as the educational datasets. Prediction of student’s performance is considered as an emerging research area under educational data mining. Previous studies have noticed that most of the available educational datasets are of small sample size. These datasets provide fewer generalization opportunities, which makes them difficult to analyze. Previous approaches use noise filtering, data balancing, GAN-based oversampling, or mostly rely on classifiers’ performance. In this paper, we proposed an approach that provides an improved model that optimizes the classifier’s performance and removes the adverse effects of noisy instances, and increase data balancing tendency in a better way. The proposed model is based on CTGAN (Conditional Tabular Generative Model), NCC (Nearest Centroid Classifier) combined with data balancing algorithm SMOTE-IPF (Iterative-Partitioning Filter) to increase dataset size by keeping their balanced nature intact and also to minimize the negative effect of noisy data points. Finally, for prediction six classifiers Random Forest (RF), Gradient Boosting (GB), CAT Boost (CT), Extra Tree (ET), K-Nearest Neighbor (KNN), and AdaBoost (AB) are used and their parameters are tuned. After parameter optimization stacking among different combination of classifiers is applied using Logistic Regression. The detailed analysis of results elaborates that the proposed model supersedes previous approaches by 2-2.5% in terms of Accuracy, and ROC.