Economic performance evaluation and classification using hybrid manifold learning and support vector machine model

Economic performance evaluation and classification is an important and challenging issue and has been gaining attention the last three decades of academic research, monetary institutions groups and business development. The purpose of this paper is to propose a hybrid model which combines support vector machine with isometric feature mapping (ISOMAP), Principal Component Analysis (PCA) and Locally Linear Embedding (LLE) utilized as a preprocessor in order to improve countries economic performance evaluation and classification capability by support vector machine. The results show that our hybrid approach SMV+ISOMAP only not has the best classification rate, but also produces the lowest incidence of Type II errors and have the excellent Receiver Operating Characteristic (ROC) curve. In addition it's capable to provide on time the economic performance classification for better investment and government decisions.