Big Data Framework Classification for Public E-Governance Using Machine Learning Techniques

Abstract

Using Machine Learning (ML) in many fields has shown remarkable results, especially in government data analysis, classification, and prediction. This technology has been applied to the National ID data (Electronic Civil Registry) (ECR). It is used in analyzing this data and creating an e-government project to join the National ID with three government departments (Military, Social Welfare, and Statistics_ Planning). The proposed system works in two parts: Online and Offline at the same time; based on five (ML) algorithms: Support Vector Machine (SVM), Decision Tree (DT), K-Nearest Neighbor (KNN), Random Forest (RF), and Naive Bayes (NB). The system offline part applies the stages of pre-processing and classification to the ECR and then predicts what government departments need in the online part. The system chooses the best classification algorithm, which shows perfect results for each government department when online communication is made between the department and the national ID. According to the simulation results of the proposed system, the accuracy of the classifications is around 100%, 99%, and 100% for the military department by the SVM classifier, the social welfare department by the RF classifier, and the statistics-planning department by the SVM classifier, respectively.