Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization
Puan Maharani Kurniawan
UIN Maulana Malik Ibrahim, Lowokwaru, Malang, 65144, Indonesia
Computer Science & IT

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Prediction of State Civil Apparatus Performance Allowances Using the Neural Network Backpropagation Method Puan Maharani Kurniawan; Agung Teguh Wibowo Almais; M. Amin Hariyadi; M. Ainul Yaqin; Suhartono Suhartono
JOIV : International Journal on Informatics Visualization Vol 7, No 3 (2023)
Publisher : Politeknik Negeri Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30630/joiv.7.3.1698

Abstract

Performance allowance is a form of appreciation given by an agency to its human resources. The Office of the Ministry of Religion of Batu City provides performance allowances to civil servants who work in the agency. Several things that affect the provision of performance allowances, such as grade, deduction, taxable income, income tax, and total tax, are used in this study to produce the total gross performance allowances and total performance allowances received. Based on the data obtained, there are some missing data from the parameters of taxable income, income tax, and total tax. This study aims to predict performance allowance when there is missing data. The method used is Neural Network Backpropagation. This study uses 480 data with split data ratios of 50:50, 60:40, 70:30, and 80:20, with epochs 40,000 and a learning rate 0,9. Four types of models used in this study are distinguished based on the number of hidden layers and epochs used. Model A uses two hidden layers to produce the highest accuracy with a 50:50 data split ratio of 65,16%. Model B uses four hidden layers to produce the highest accuracy with a 50:50 data split ratio of 69,34%. Model C uses six hidden layers to produce the highest accuracy with a 50:50 data split ratio of 68,18%. Model D uses eight hidden layers to produce the highest accuracy with a 50:50 data split ratio of 70,90%.
Co-Authors Agung Teguh Wibowo Almais M. Ainul Yaqin M. Amin Hariyadi Suhartono Suhartono