LeetCode Contest Analytics

Overview

• A data engineering project to extract, transform, and load LeetCode contest ranking data and contest problems data.
• This project was created to analyze data from past LeetCode contests and top contestants, aiming to set the basis for efficient contest practice and further data analytics from each user profile.
• Data were collected from the top 50,000 users of LeetCode global contest ranking and 2334 problems of 585 contests.
• Some analytics with visualizations are provided in analytics/analytics.md.
• The collected data are stored in the data/ folder.

Technology Stack

• Orchestration: Apache Airflow
• Storage: Amazon S3
• ETL jobs: Pandas, AWS Glue
• Data warehouse & Analytics: Amazon Redshift
• For additional libraries, please consult requirements.txt.

Potential Ideas

• Machine Learning/Deep Learning models.
• User profile analysis and ranking prediction.
• Further analytics and statistics from other LeetCode data.
• Data pipelines and analytics for other platforms.