Abstract: This study presents our investigation on different hybrid classification models which integrate support vector machines with nearest neighbor algorithm. We study the advantages and disadvantages of Support Vector Machines (SVM) classification and k-nearest neighbor (KNN) classification in performing their classification tasks. In our investigation, we found that the well-performing SVM classification approach may suffer from high time consumption, high CPU and physical memory usages, due to its convoluted training and classifying processes, especially when the dimensionality of data is high. On the other hand, KNN classification approach which implements NN algorithm is outstanding with its simplicity and low cost training process. However, it has been reported to be less accurate than the SVM classification. Many research works have been carried out in order to further improve the performance of the established SVM classifier by integrating Nearest Neighbor (NN) algorithm into the conventional SVM classification approach. The research works which have been reviewed and investigated in this paper emphasize in simplifying the convoluted training and classifying processes and further improving the classification algorithm of the conventional SVM, using NN algorithm. Overall, we concluded that while SVM classification approach has been reported as one of the best-performing classifiers since some decades now, it could be further improved by using NN algorithm in order to obtain more effective and efficient classification models.