Plant Disease Detector
Detect plant diseases early from symptom descriptions using advanced natural language processing
Multinomial Naive BayesAccuracy: 100%F1 Score: 1.00
Symptom Description
Describe the symptoms you observe in your plants
๐ How It Works
Our disease detector uses Multinomial Naive Bayes with Laplace smoothing (ฮฑ=1.0) combined with TF-IDF text vectorizationโa probabilistic NLP algorithm that analyzes symptom descriptions to identify disease patterns with perfect 100% accuracy on 1,600 Philippine crop disease reports.
๐ What to Describe:
- ๐Leaf Changes: Yellow/brown spots, discoloration, curling edges
- ๐ฑGrowth Issues: Stunted growth, wilting, weak stems
- ๐ฆ Disease Signs: Powder, mold, lesions, unusual substances
- ๐Extent: How much of the plant is affected
๐ Model Performance:
Accuracy
100%
Precision
1.00
F1 Score
1.00
โ Trained on 1,600 disease symptom descriptions
โ TF-IDF vectorization + Naive Bayes (ฮฑ=1.0 smoothing)
โ Covers rice, corn, sugarcane, and vegetable diseases