Tourism Package Prediction Model
Best Model
Gradient Boosting
Model Performance
| Metric | Value |
|---|
| Accuracy | 0.9203 |
| Precision | 0.9174 |
| Recall | 0.6452 |
| F1 Score | 0.7576 |
Best Hyperparameters
{'classifier__n_estimators': 100, 'classifier__max_depth': 7, 'classifier__learning_rate': 0.1}
Features
The model predicts whether a customer is likely to purchase a Wellness Tourism Package using customer demographic, travel, and behavioural attributes.
Preprocessing Pipeline
The model was trained using an Scikit-learn Pipeline consisting of:
- Missing Value Imputation
- One-Hot Encoding for categorical variables
- Standard Scaling for numerical variables
- Gradient Boosting Classifier
MLOps Workflow
- Dataset registered in Hugging Face Dataset Hub
- Model trained using Hyperparameter Tuning (RandomizedSearchCV)
- Best model selected based on evaluation metrics
- Model registered in Hugging Face Model Hub
- Ready for deployment in Hugging Face Spaces
