Data analysis, visualization, and prediction on NYC Airbnb data
In this project we analyze Airbnb data gathered from the New York City area in 2019 and create various models to predict future AirBnb prices.
This datset contains 48,895 entries and 16 columns of qualitative and quantitative data. Our models consist of 34,218 entries for training and 14,666 for testing.
| Model | RSME Average | 10-Fold RSME Average |
|---|---|---|
| Linear Regression | 0.495 | 0.498 |
| Random Forest Regressor | 0.493 | 0.509 |
| XgBoost | 0.473 | NaN |
| Lasso Regression | 0.693 | 0.698 |
| Model | Accuracy |
|---|---|
| Random Forest Classifier | 83.4% |
| Logistic Regression | 84.5% (Overall Best) |
| Decision Tree | 80.9% |
| KNN | 84.1% (when k = 15) |
| NYC Borough | Average Listing price (USD) |
|---|---|
| Manhattan | $291.38 |
| Staten Island | $277.62 |
| Brooklyn | $186.87 |
| Queens | $167.10 |
| Bronx | $106.70 |
- Python
- Sklearn
- Numpy
- Matplotlib
- Pandas
- Jupyter Notebook