Building a Classification and Forecasting Model to Support Decision Making

Abstract

Data might be one of the most valuable assets of any corporation but only if it knows how to reveal valuable knowledge hidden in raw data. Data mining allows extracting precious knowledge from historical data, and predicting the outcomes of future situations. A large database of about thousands is stored in an Oracle database system. It contains the demographical information and more data about the history of the pensioner financial history at the National Pension and Social Insurance Fund. Managing loans for regular use or investment is unpredicted. The main aim of this research is to predict the numbers of borrowers and the budget of investment & loans by using classification and regression techniques. The model was built by using the Trees function-RJ48 algorithm. Different regression algorithms were used such as Sequential Minimal Optimization Regression, Linear Regression, and Simple Linear Regression. Regression results were compared against the actual money spent for the years 2017-2019 to predict the number of borrowers and the budget. The simple linear regression proved to be the most accurate, with the least error ratio.