---
language: en
license: mit
library_name: scikit-learn
tags:
- travel
- destination-prediction
- clustering
- recommendation-system
---

# Destination Cluster Predictor

## Model Description

This model is a machine learning system designed to predict and recommend travel destinations based on user preferences and requirements. It uses a combination of clustering and classification techniques to group similar destinations and make personalized recommendations.

### Model Type

The model consists of three main components:
- A clustering model (`destination_clustering_model.pkl`)
- Label encoders for categorical features (`destination_label_encoders.pkl`)
- A scaler for numerical features (`destination_scaler.pkl`)

### Input Features

The model takes the following input features:

1. **Interest**: Combinations of interests (Mountains, Wildlife, Adventure, Culture, etc.)
2. **Goal**: Travel goals (Adventure, Exploration, Photography, Trekking, etc.)
3. **Climate**: Weather conditions (Temperate, Cold, Moderate, Cool, Warm, etc.)
4. **Solo/Group**: Travel type (Solo, Group, or Solo/Group)
5. **Access**: Transportation options (Road, Trek, Air, Boat, etc.)
6. **Distance**: Numerical value (10-1500 km)
7. **Latitude**: Numerical value (24-37)
8. **Longitude**: Numerical value (60-78)
9. **Activity**: Various activities and their combinations

### Output

The model outputs:
- A predicted destination cluster
- Top 5 destination recommendations based on the input preferences

## Training Data

The model was trained on a dataset of travel destinations with their associated features and characteristics. The training data is stored in `data.xlsx` and contains 125 entries.

## Training Procedure

The model uses a combination of:
- Label encoding for categorical variables
- Standard scaling for numerical features
- Clustering algorithm for destination grouping

## Evaluation

The model's performance is evaluated based on:
- Cluster coherence
- Recommendation relevance
- User preference matching

## Limitations

- The model's recommendations are limited to the destinations present in the training data
- Geographic coordinates are constrained to specific ranges (Latitude: 24-37, Longitude: 60-78)
- Distance recommendations are limited to 10-1500 km range

## Usage

```python
# Example usage
from predictor.models import DestinationPredictor

predictor = DestinationPredictor()
recommendations = predictor.predict(
    interest="Mountains",
    goal="Adventure",
    climate="Temperate",
    travel_type="Solo",
    access="Road",
    distance=500,
    latitude=30,
    longitude=70,
    activity="Trekking"
)
```

## Environmental Impact

The model is lightweight and can run efficiently on standard hardware. No special GPU requirements are needed for inference.

## Citation

If you use this model in your research or application, please cite:

```bibtex
@misc{destination_predictor,
  author = {Your Name},
  title = {Destination Cluster Predictor},
  year = {2024},
  publisher = {Hugging Face},
  journal = {Hugging Face Hub},
  howpublished = {\url{https://huggingface.co/your-username/destination-predictor}}
}
```

## License

This model is licensed under the MIT License.