Klasifikasi Kualitas Karet Menggunakan Metode Random Forest Pada Perusahaan XYZ

Company XYZ, a crumb rubber processor in Sumatra, Indonesia, assigns each production lot a Low, Medium, or High quality category by laboratory testing, but recorded parameters are not linked to them. This study tests how well Random Forest reproduces the categories from six parameters in 1,031 lots, only 13.5% of which are Low. Random Forest was compared with Support Vector Machine (SVM), Decision Tree, and two naive baselines using hold-out, repeated, and nested cross-validation. Cross-validated macro-F1 of the tuned Random Forest was 0.409, higher than Decision Tree (0.365; p = 0.028) and not significantly different from SVM (0.385; p = 0.13), yet only slightly above random guessing (0.330). No model exceeded the accuracy of always guessing High (0.507), and the Low class fared worst (F1 about 0.23). The six parameters barely differ between classes, so the data carry little class information. The model suits only flagging lots for laboratory re-testing.

Authors

Institutions

Publication Details

Journal
Jurnal Manajemen dan Teknologi
Published
2026-11-18
DOI
https://doi.org/10.63447/jmt.v3i2.2195
Primary Topic
Data Mining and Machine Learning Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Klasifikasi Kualitas Karet Menggunakan Metode Random Forest Pada Perusahaan XYZ

Hardini Novianti, Ken Ditha Tania, Mira Afrina, Alsella Meiriza et al.
Jurnal Manajemen dan Teknologi
Data Mining and Machine Learning Applications
article

Klasifikasi Kualitas Karet Menggunakan Metode Random Forest Pada Perusahaan XYZ

Hardini Novianti, Ken Ditha Tania, Mira Afrina, Alsella Meiriza, Ongki Juliandra
article en

Abstract

Company XYZ, a crumb rubber processor in Sumatra, Indonesia, assigns each production lot a Low, Medium, or High quality category by laboratory testing, but recorded parameters are not linked to them. This study tests how well Random Forest reproduces the categories from six parameters in 1,031 lots, only 13.5% of which are Low. Random Forest was compared with Support Vector Machine (SVM), Decision Tree, and two naive baselines using hold-out, repeated, and nested cross-validation. Cross-validated macro-F1 of the tuned Random Forest was 0.409, higher than Decision Tree (0.365; p = 0.028) and not significantly different from SVM (0.385; p = 0.13), yet only slightly above random guessing (0.330). No model exceeded the accuracy of always guessing High (0.507), and the Low class fared worst (F1 about 0.23). The six parameters barely differ between classes, so the data carry little class information. The model suits only flagging lots for laboratory re-testing.

Jurnal Manajemen dan TeknologiVol. 3(2)
Sriwijaya University (ID)
Data Mining and Machine Learning Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon