Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JINAV: Journal of Information and Visualization
Bunga, Yhegi Rombe
Unknown Affiliation
Computer Science & IT Decision Sciences, Operations Research & Management Engineering Library & Information Science Mathematics

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

Found 1 Documents
Search

 1 
Rainfall Classification Using Output Statistics Models Based on Classification and Regression Trees with Principal Component Analysis Preprocessing Rais, Zulkifli; Hafid, Hardianti; Bunga, Yhegi Rombe
JINAV: Journal of Information and Visualization Vol. 7 No. 1 (2026)
Publisher : PT Mattawang Mediatama Solution

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Makassar City has a varied monsoon rainfall pattern, so rainfall prediction is an important challenge in disaster mitigation and resource management. Data mining techniques such as classification with the Classification and Regression Trees (CART) algorithm can be used to classify rainfall and analyze historical data, but the risk of overfitting high-dimensional data requires dimension reduction such as Principal Component Analysis (PCA). To improve accuracy, the Output Statistics Model (MOS) approach that combines numerical data and observations is also used. The results of dimension reduction using the Principal Component Analysis (PCA) method showed that of the initial seven variables, only three main components (, , and ) were retained because they had eigenvalues greater than 1 and were able to explain the data variance significantly. The decision tree model that was formed resulted in an accuracy rate of 72.34% in training data. Where the model can classify most of the training data into the correct rainfall category. In the data testing, the model was able to achieve an accuracy level of 71.43%, which shows that the model has good generalization ability to new data and does not experience overfitting.
Co-Authors Hardianti Hafid Rais, Zulkifli