- Build a functional eBay web scraping tool.
- Store and maintain a PostgreSQL database.
- Create a live dashboard that pulls from the database and various metrics using the data.
With the abundance in the number of phones listed on online marketplaces, it overwhelms the user in selecting satisfactory auctions. My approach includes building a web-scraping tool to collect data on phone auctions and calculating various metrics to aid in auction selection.
- Create the web-scraping script.
- Containerize the script.
- Utilize workflow services to execute scripts.
- Execute scraping to get more auctions.
- Execute the removal of duplicate ids.
- Monitor the number of rows in the database and empty if it reaches over 10,000 samples (our database storage limit)
- Check if an auction is active.
- Store the data into a database.
- Preprocess the data.
- Perform exploratory data analysis.
- Deployment
- Docker
- Python
- Apache Airflow
- PostgreSQL
- AWS