relation: https://khub.utp.edu.my/scholars/17072/
title: Revenue Prediction for Malaysian Federal Government Using Machine Learning Technique
creator: Noor, N.
creator: Sarlan, A.
creator: Aziz, N.
description: Every country has its own federal government. Each federal government will have its own financial account which consist of revenue and expenditure. Focusing on the revenue, it has many sources that includes three main categories. They are tax revenue, non-tax revenue and non-revenue receipts. The revenue will then be used for operational and development purposes. Currently in Malaysia, the federal government revenue is only using forecasting. This can cause large forecasting error. Though it can be overcome using predictive analytics. Since there are many machine learning methods available, the appropriate methods can be identified to do the prediction. Based on previous research, feed forward neural network (FFNN), random forest and linear regression seems to be the most suitable. After conducting several experiments, it is found that FFNN achieved highest accuracy, followed by random forest. As for linear regression, it does not achieve good accuracy, thus it is considered as a not suitable method to be used on the federal government revenue dataset. © 2022 ACM.
publisher: Association for Computing Machinery
date: 2022
type: Conference or Workshop Item
type: PeerReviewed
identifier:   Noor, N. and Sarlan, A. and Aziz, N.  (2022) Revenue Prediction for Malaysian Federal Government Using Machine Learning Technique.  In: UNSPECIFIED.     
relation: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85132290252&doi=10.1145%2f3524304.3524337&partnerID=40&md5=18d2b239ebcf3b6cd7b3e8b589134926
relation: 10.1145/3524304.3524337
identifier: 10.1145/3524304.3524337