Rainfall Prediction Using Random Forest
Project Overview This project predicts whether it will rain tomorrow using weather observations from the Rain in Australia dataset.
Dataset
- Rain in Australia Dataset
- 142,193 records after preprocessing
Features Used
- MinTemp
- MaxTemp
- Rainfall
- Humidity9am
- Humidity3pm
- RainToday
Machine Learning Model
- Random Forest Classifier
Results
- Accuracy: 82.72%
Confusion Matrix
| Actual / Predicted | No Rain | Rain |
|---|---|---|
| No Rain | 20796 | 1302 |
| Rain | 3611 | 2730 |
Feature Importance
- Humidity3pm
- MinTemp
- MaxTemp
- Humidity9am
- Rainfall
- RainToday
Tools Used
- Python
- Pandas
- Scikit-Learn
- Matplotlib
- Jupyter Notebook