Estimation of the Country Ranking Scores on the Global Innovation Index 2016 Using the Artificial Neural Network Method

Abstract

The Global Innovation Index (GII) aims to rank countries using different innovation factors. This ranking list enables countries to observe their potential status according to the rankings of other countries. The countries are classified under four groups according to the World Bank Income Group Classification on the GII list. The groups are named as; low income (LI), lower-middle income (LM), upper-middle income (UM) and high income (HI). Also, every country has a score in this ranking list. In this study, the ranking scores of 128 countries are estimated using the artificial neural network (ANN). We chose the relevant 27 features on GII 2016 Report, as input data. The significance of this paper is that; it is the first curve fitting and estimation of the score processes on GII 2016 dataset. The low root mean square error (RMSE) value which is obtained in an experimental study shows that the fitting structure is good enough to determine the approximate score of the countries in GII list. The results also show that the selected 27 features are sufficient for obtaining the income score of the countries. Increasing the number of features would lower the RMSE value and enable better approximation in the curve fitting process. The final results can assist the countries in achieving long-term output growth and improving their innovation capabilities.