This notebook contains two machine learning projects: Flight Price Prediction (Regression) and Customer Satisfaction Prediction (Classification). Both projects are designed to predict flight prices and customer satisfaction levels using various machine learning algorithms. The models are deployed using Streamlit and experiment tracking is managed using MLflow.
Predict flight ticket prices based on multiple factors such as departure time, source, destination, and airline type.
- Python, Streamlit, Machine Learning, MLflow, Data Analysis
Travel and Tourism
- Load and preprocess flight price data.
- Perform exploratory data analysis (EDA) to identify trends and correlations.
- Train regression models like Linear Regression, Random Forest, and XGBoost.
- Develop a Streamlit app to predict flight prices based on user input (e.g., route, time, and date).
- Use MLflow for tracking experiments and saving models.
Predict customer satisfaction levels based on features such as customer feedback, demographics, and service ratings.
- Python, Streamlit, Machine Learning, MLflow, Data Analysis
Customer Experience
- Load and preprocess customer satisfaction data.
- Perform exploratory data analysis (EDA) to understand relationships between features.
- Train classification models like Logistic Regression, Random Forest, and Gradient Boosting.
- Develop a Streamlit app to predict customer satisfaction levels based on input features.
- Use MLflow for tracking experiments and saving models.
- For Travelers: Help travelers plan trips by predicting flight prices based on preferences (route, time, etc.).
- For Travel Agencies: Assist in price optimization and marketing strategies.
- For Businesses: Enable businesses to budget for employee travel by forecasting ticket prices.
- For Airlines: Support airlines in identifying trends and optimizing pricing strategies.
- For Airlines: Enhance customer experience by predicting and addressing dissatisfaction.
- For Businesses: Provide actionable insights to improve services.
- For Marketing: Identify target customer groups for specific promotions.
- For Management: Support decision-making for customer retention strategies.
First, install the required dependencies using the following command:
pip install -r requirements.txt
To run the Streamlit app, use the command below. This will launch a web interface where you can interact with both projects.
streamlit run app.py- Users can input parameters like route, time, and date to get predicted flight prices.
- Displays visualizations of flight price trends.
- Uses regression models for price predictions.
- Users can input passenger features such as age, type of travel, and service ratings to predict satisfaction levels.
- Visualizes customer satisfaction trends.
- Uses classification models for satisfaction predictions.
Both projects utilize MLflow to track and manage experiments, models, and metrics:
- MLflow logs parameters, metrics (e.g., accuracy, RMSE), and artifacts (e.g., model files, visualizations).
- All trained models are saved and organized using MLflow's model registry.
- For data preprocessing, model training, and MLflow integration.
- Cleaned and processed CSV files containing flight price and customer satisfaction data.
- Regression and classification models trained and logged with MLflow.
- Interactive app for data visualization and predictions, integrated with MLflow metadata.
- Comprehensive documentation covering methodology, analysis, and insights.
- Python 3.8+
- Columns:
- Airline, Date_of_Journey, Source, Destination, Route, Dep_Time, Arrival_Time, Duration, Total_Stops, Additional_Info.
- Columns:
- Gender, Customer Type, Age, Type of Travel, Class, Flight Distance, Inflight Wifi Service, Seat Comfort, On-board Service, Cleanliness, Departure Delay, Arrival Delay, Satisfaction.
This project is licensed under the MIT License - see the LICENSE file for details.