Modelling of Sudan’s Economy Composition using Machine Leaming Approaches

Modelling of economy systems is critical as it provides means of addressing and future forecasting of economic outcomes, hence accurate models should be developed. The purpose of this research is to show the effectiveness of Machine Learning approaches in modelling of economy systems. The economy of Sudan is used to reflect this objective, taking the Gross Domestic Product (GDP) as a monetary measure of economy. It was found that Agriculture, Industry and Service are the main contributors of Sudan’s GDP in accordance to the GDP data of Sudan from 1960 to 2016. In this research, Regression and Support Vector Machine algorithms were used to develop two different models of Sudan’s economy. Both models have produced satisfying results in predicting, however the SVM model has out-performed the Polynomial Regression model.