Crop Recommendation System
Karen Nam, Salahuddin Syed, Seonyeong Heo,
Tumi Samuel-Ipaye, Yein Choi
Introduction
●Focus: Optimize agricultural productivity
●Align crops with regional climatic conditions
●Benefits:
○Increased crop yield
○Improved resource utilization
○Reduced risk of crop failure
○Enhanced profitability
○Sustainable farming practices
https://cdn.dribbble.com/users/1034323/screenshots/16279921/media/6172d47ca541d92ef9279ceb59d359cb.gif
●Soil components
○Nitrogen (N)
○Phosphorus (P)
○Potassium (K)
○pH
●Environmental factors
○Temperature
○Humidity
○Rainfall
Data Description
Data Description
●Output labels (21 -> 10)
Raw (label numbers = 21) Subset (label numbers = 10)
WORKFLOW
Full Dataset
(n = 1000)
Data Preprocessing Data Splitting
Training
(70%)
Test
(30%)
Model Training
Precision (average = ‘macro’)
10-fold CV
Decision Tree
Model Evaluation
Accuracy
Standardization
Label encoding
Precision
Recall
Random Forest
kNN
Logistic Regression
SVM
Stratified data split
Confusion Matrix
kNN LR, DT, RF, SVM
Model Accuracy Precision Recall
Decision Tree 1.000 1.000 1.000
Random Forest 1.000 1.000 1.000
Logistic Regression 1.000 1.000 1.000
SVM 1.000 1.000 1.000
kNN 0.997 0.997 0.997
Model Comparison Results
Example 1 - Hot and humid environment
Example 2 - Warm and moderately humid environment
Crop Recommendation
