Case file CS-02SYS751Group project
Smart Crop Recommendation System: Complementing Sistem GeoTanaman by Jabatan Pertanian Malaysia
A decision-support system that recommends crops with machine learning, designed to complement Sistem GeoTanaman by Jabatan Pertanian Malaysia (Department of Agriculture).
- Headline
- 95.6%accuracy, Random Forest
- Course
- SYS751 Advanced Decision Support Systems
- Format
- Group project
- Date
- July 2025
- Programme
- Master of Data Science, UiTM
- Team
- 3 members
// 01 · Problem
The problem
Sistem GeoTanaman is a system by Jabatan Pertanian Malaysia, the Department of Agriculture. The project identified two gaps in the existing system: its advice is generic, and it relies on outdated soil data.
The team set out to build a smart crop recommendation system that complements GeoTanaman instead of replacing it.
// 02 · Approach
The approach
STEP 01
Gap analysis
Reviewed the existing Sistem GeoTanaman and framed the project around generic advice and outdated soil data.
STEP 02
Model building
Trained and compared crop-recommendation classifiers in Python with scikit-learn and XGBoost.
STEP 03
Decision-support app
Wrapped the model in a Streamlit app and presented it as a live demo in July 2025.
// 03 · Results
The results
95.6%
Accuracy
Random Forest
3
Team members
Live
Streamlit demo
July 2025
- Random Forest reached 95.6% accuracy.
- The system was demonstrated live as a Streamlit application.
// 04 · Tools
Tools and methods
- Python
- scikit-learn
- XGBoost
- Random Forest
- Streamlit
// 05 · Role & team
Role and team
Group project for SYS751 Advanced Decision Support Systems, Master of Data Science, UiTM.
- Adni Muhammad Fahmi bin Mashudi
- Muhammad Aiman Akmal bin Pardan
- Nabil TakiyuddinThis site