https://github.com/nafishr24/flight-prediction-kaggle

This project predicts ticket types for airlines, including economy, premium economy, business, and first class, using machine learning to improve ticket classification and enhance the booking experience.
https://github.com/nafishr24/flight-prediction-kaggle

data-science machine-learning python

Last synced: 3 months ago
JSON representation

This project predicts ticket types for airlines, including economy, premium economy, business, and first class, using machine learning to improve ticket classification and enhance the booking experience.

• Host: GitHub
• URL: https://github.com/nafishr24/flight-prediction-kaggle
• Owner: nafishr24
• License: mit
• Created: 2025-04-01T11:58:37.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-04-19T13:58:57.000Z (over 1 year ago)
• Last Synced: 2025-07-03T15:56:20.259Z (about 1 year ago)
• Topics: data-science, machine-learning, python
• Language: Jupyter Notebook
• Homepage: https://www.kaggle.com/datasets/iamavyukt/goibibo-flight-data
• Size: 1.13 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Flight Ticket Type Prediction

### Description

This project predicts ticket types for airlines, including economy, premium economy, business, and first class, using machine learning to improve ticket classification and enhance the booking experience.

The dataset used contains detailed information on domestic flights in India for the period of July to August 2023. The dataset can be downloaded from [Kaggle - Goibibo Flight Data](https://www.kaggle.com/datasets/iamavyukt/goibibo-flight-data).

#### Dataset Details:

This file contains detailed information about domestic flights in India for the period of July to August 2023.

**Columns:**

- **flight_date:** The date of the flight in YYYY-MM-DD format.

- **airline:** Name of the airline operating the flight.

- **flight_num:** The flight number.

- **class:** Travel class (e.g., economy, business).

- **from:** Departure city.

- **dep_time:** Scheduled departure time.

- **to:** Arrival city.

- **arr_time:** Scheduled arrival time.

- **duration:** Total duration of the flight in hours and minutes.

- **price:** Ticket price in Indian Rupees (INR).

- **stops:** Number of stops (e.g., non-stop, 1-stop).

The dataset is structured to facilitate analysis of flight durations, prices, and other travel patterns, making it suitable for research and exploration of the Indian domestic aviation market.

### Methods Used

This project employs **KMeans Clustering** to group flights based on key features, with the optimal number of clusters determined using the **Elbow Method**.

### Models Used

Several machine learning models were tested and compared:

- **Logistic Regression**

- **Decision Tree**

- **Random Forest**

- **K-Nearest Neighbors (KNN)**

- **XGBoost**

Each model was evaluated under two scenarios:

1. **Without tuning** (using default parameters)

2. **With tuning using GridSearchCV** (to find the best parameter combination)

### Results & Conclusion

- **XGBoost showed performance improvement after tuning**, making it the best model for flight ticket classification.

- **Random Forest and KNN remained stable but require improved recall for minority classes.**

- **Decision Tree experienced a performance drop after tuning.**

- **Logistic Regression remains suboptimal for this dataset.**

➡ **XGBoost is the best model after tuning but requires further optimization to improve recall for minority classes.**

➡ **Random Forest and KNN can still be considered with additional tuning.**

### Social Media

Follow me on:

- [Instagram](https://www.instagram.com/nafishusen24/)

- [LinkedIn](https://www.linkedin.com/in/nafishusen24/)

- [GitHub](https://github.com/nafishr24/)