This is the capstone project for the University of San Diego’s ADS 599 Capstone Project course.
This project tackles issues of airfare uncertainty and limited filtering options in flight aggregation apps. We focus on two main objectives: predicting airfare changes over time and correlating airfare prices with desired amenities. For the first objective, we assist customers in predicting price changes within one day, one week, and one month. This empowers travelers with insights for making informed decisions on when to book flights. The second objective addresses the challenge of limited filtering options for specific features. Leveraging machine learning, our model helps users predict how airfare prices correlate with desired amenities, providing a more comprehensive understanding of the relationship between ticket prices and specific features.
- Data Extraction: Alternative method that details the steps taken to extract and store data from BFM API.
- Data Preparation: Details the steps taken to prepare and preprocess the raw data for analysis.
- Data Exploration: The exploratory data analysis (EDA) phase is documented. It includes visualizations, statistical analyses, and insights gained from exploring the dataset.
- Modeling: Application of eleven machine learning models to the preprocessed data. It covers model training, evaluation, and tuning processes. The results and performance metrics of each model are presented, providing a comprehensive overview of the modeling phase of the capstone project.
- Notebook: Contains the contents from all other Notebooks.
- Visualizations: Contains visualizations derived within the Notebook.
- Data: The ZIP file comprises the primary data utilized in the project, sourced from Sabre's Bargain Finder Max API.
- Article: Article of the study.
- Presentation: Presentation of the study.
- Python Library: List of the libraries used.
- Python