Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pembelajaran dan Biologi Nukleus
Muhammad Afifi Andriansyah
Universitas Negeri Surabaya
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Education Environmental Science Immunology & microbiology Languange, Linguistic, Communication & Media

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

Found 1 Documents
Search

 1 
Detection of Soil Organic Matter Using IoT-Based Soil Color Sensors with Random Forest Method Muhammad Afifi Andriansyah; Lusia Rakhmawati; I Gusti Putu Asto Buditjahjanto
JURNAL PEMBELAJARAN DAN BIOLOGI NUKLEUS Vol 12, No 1: Jurnal Pembelajaran Dan Biologi Nukleus March 2026
Publisher : Universitas Labuhanbatu

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36987/jpbn.v12i1.9077

Abstract

Background: Soil organic matter (SOM) is an important indicator of soil fertility that plays a role in agricultural productivity and ecosystem sustainability. However, conventional laboratory-based methods still have limitations in terms of time, cost, and do not support real-time monitoring. Therefore, an approach based on sensors and machine learning is needed for quick and efficient estimation. This study proposes an Internet of Things (IoT)-based system that integrates an RGB soil color sensor (TCS3200) and a pH sensor to estimate soil organic matter content using a Random Forest algorithm. Methodology: Laboratory analysis was conducted using the Walkley–Black method. Soil samples were taken from seven locations (T1–T7). The Random Forest model was developed with parameters n_estimators = 100 and max_depth = 10, and validated using a train-test split method (80:20). Findings: The results showed that darker-colored soils have higher organic carbon content. The model's error values indicated an MAE of 0.031 and an RMSE of 0.032. The Random Forest model achieved a classification accuracy of 85.7% and a coefficient of determination R² ≈ 0.97. Contributions: This study contributes by developing an integrated IoT and machine learning system capable of quickly, accurately, and cost-effectively estimating soil organic matter to support precision agriculture
Co-Authors I Gusti Putu Asto Buditjahjanto Lusia Rakhmawati