Implementation

1 . Environment Setup: Install and import the required PennyLane and classical ML libraries needed to run the quantum and classical models.

2. Library Imports: Import all necessary modules: Scikit-learn for classical ML, PennyLane for quantum conputing, as well as standard data science libraries.

3.Data Loading: Mount Google Drive and load the Air Quality & Mobility dataset into a pandas DataFrame.

3.Data Cleaning: Remove non-numerical and irrelevant columns (Date, City) that won't contribute to model training.

4.Data Labeling: Convert continuous AQI values into binary classes (Acceptable / Unhealthy), then split the data into features and target.

6. Train/Test Split: Divide the dataset into 80% training and 20% testing subsets.

6. Feature Scaling: Standardize all features to zero mean and unit variance.

8. Feature Selection: Use a Random Forest to rank feature importance and select the top 3 most predictive features, enabling a fair comparison with the 3-qubit quantum model.

9. Classical Model (Random Forest): Train and evaluate a Random Forest classifier on the 3 selected features as the classical baseline.

10. Quantum Data Preparation: Assign the scaled 3-feature subsets as the input data for the quantum model.

11. Label Encoding: Map class labels to binary integers (0/1) and prepare training samples for quantum-compatible input.

12. Quantum Model Training (VQC): Define and train a Variational Quantum Classifier using PennyLane's angle embedding and strongly entangling layers, optimized with the Adam optimizer over 60 epochs (entire training sample).

13. Quantum Model Evaluation: Run the trained quantum circuit on test data and evaluate classification performance using accuracy and F1-score.

13. Quantum Model Evaluation: Run the trained quantum circuit on test data and evaluate classification performance using accuracy and F1-score.

14. Hybrid Quantum-Classical Model: Use a fixed quantum circuit as a feature extractor, concatenate the quantum features with the original classical features, then train a Random Forest on the augmented data.

14. Hybrid Quantum-Classical Model: Use a fixed quantum circuit as a feature extractor, concatenate the quantum features with the original classical features, then train a Random Forest on the augmented data.

15. Results Comparison: Compare the accuracy of all three models (Random Forest, VQC, and Hybrid) and visualize the results in a bar chart using Matplotlib.