Perbandingan Algoritma C5.0 Dan Random Forest Untuk Klasifikasi Calon Penerima Program Keluarga Harapan
Dublin Core
Title
Perbandingan Algoritma C5.0 Dan Random Forest Untuk Klasifikasi Calon Penerima Program Keluarga Harapan
Subject
Algoritma C5.0, Random Forest, Klasifikasi
Description
he Family Hope Program (PKH) is a conditional cash transfer program aimed at improving the welfare of poor households in Indonesia. However, in practice, problems related to targeting accuracy still occur, as manual data collection often leads to errors and potential social inequality. Therefore, a technology-based method is required to assist in the classification of beneficiaries more objectively and accurately.
This study aims to compare the performance of the C5.0 and Random Forest algorithms in classifying prospective PKH recipients based on population data in Bakealu Village, South Wakorumba District, Muna Regency. The C5.0 algorithm was chosen for its ability to generate simple and interpretable decision trees, while Random Forest was selected due to its stability and high prediction accuracy on complex datasets.
The results of this study show that the developed application achieved a UAT score of 85.27%, which falls into the “Very Good” category, indicating that the system is feasible and easy for users to operate. In the evaluation of model performance using the confusion matrix, the Random Forest algorithm demonstrated superior results with an accuracy of 84.38%, precision of 100%, recall of 84.38%, and an F1-Score of 91.53%, while the C5.0 algorithm achieved an accuracy of 81.25%, precision of 100%, recall of 81.25%, and an F1-Score of 89.65%. These findings indicate that Random Forest is more suitable for achieving higher predictive accuracy, whereas C5.0 remains beneficial when a more interpretable and transparent model is needed, allowing both algorithms to be applied appropriately in supporting more accurate targeting of PKH beneficiaries.
This study aims to compare the performance of the C5.0 and Random Forest algorithms in classifying prospective PKH recipients based on population data in Bakealu Village, South Wakorumba District, Muna Regency. The C5.0 algorithm was chosen for its ability to generate simple and interpretable decision trees, while Random Forest was selected due to its stability and high prediction accuracy on complex datasets.
The results of this study show that the developed application achieved a UAT score of 85.27%, which falls into the “Very Good” category, indicating that the system is feasible and easy for users to operate. In the evaluation of model performance using the confusion matrix, the Random Forest algorithm demonstrated superior results with an accuracy of 84.38%, precision of 100%, recall of 84.38%, and an F1-Score of 91.53%, while the C5.0 algorithm achieved an accuracy of 81.25%, precision of 100%, recall of 81.25%, and an F1-Score of 89.65%. These findings indicate that Random Forest is more suitable for achieving higher predictive accuracy, whereas C5.0 remains beneficial when a more interpretable and transparent model is needed, allowing both algorithms to be applied appropriately in supporting more accurate targeting of PKH beneficiaries.
Creator
Putri Elisya, Asrul Sani, Asa Hari Wibowo
Source
https://animator.uho.ac.id/index.php/journal/article/view/1283
Publisher
Informatics Engineering Department of Halu Oleo University
Date
2025-12-26
Contributor
Sri Wahyuni
Rights
ISSN : 3030-9735
Format
PNG
Language
English
Type
Text
Files
Collection
Citation
Putri Elisya, Asrul Sani, Asa Hari Wibowo, “Perbandingan Algoritma C5.0 Dan Random Forest Untuk Klasifikasi Calon Penerima Program Keluarga Harapan,” Repository Horizon University Indonesia, accessed January 11, 2026, https://repository.horizon.ac.id/items/show/9923.