Estimation of global innovation index scores with machine learning

In today's dynamic global economy, innovation plays a critical role in driving growth and success. Measuring innovation is challenging, but the Global Innovation Index (GII) is a valuable tool identifying essential indicators of innovation. However, the process of calculating the GII is complex and time-consuming, which poses a significant challenge. To streamline the process, this study focused on reducing the number of parameters required to calculate the GII using a feature selection method. The study used an artificial neural network (ANN) methods and various regression methods to estimate GII scores for 125 countries from 2012-2022. The study's findings suggest that the selected 38 parameters were sufficient to accurately calculate the GII score, and the ANN model was the most precise method. A prototype interface was designed to predict GII scores, allowing for easier and more efficient estimation. This study makes an important contribution to the research literature by reducing the parameter count and improving estimation accuracy, which could simplify the process of calculating the GII and make it more accessible to organizations and policymakers worldwide.