PokeCardeDex is a one-stop web application built for Pokemon TCG collector's. The platform integrates organizing portfolios, warehouse for pokemon card data, price monitoring, data analytics, and AI-driven insights** to increase user agency and decision-making.
- Frontend: Next.JS written with Javascript and Typescript.
- Backend: FastAPI with MongoDB storage
- Deployment: Aiming to be hosted on AWS cloud services for scalability and reliability.
- Dashboard displaying Pokemon cards from each set in an easily identifiable manner.
- Filters for setting and organizing sets to easily identify your specific chase cards.
- Integration with price monitoring to make better insights into whether a specific booster pack or box is worth a certain price depending on the Chase card's prices.
- Web scraping tools are used to monitor multiple websites for certain cards or products and will update every 15 minutes for API request optimization.
- Why every 15 minutes? this is because Pokemon cards generally do not have high volatility and would optimise API fetch calls allowing much faster rendering with less overhead.
- Statistical formulas to gauge expected returns of specific goods depending on current market price and card set.
- Risk analysis to provide a gauge of how much risk the user is willing to take and when a price point is too high.
- Model training Identify keep topics and regions to ingest data in which can provide a relationship into speculating card prices.
- Per set release quarterly job
- Updates the current set list every quarter for new sets.
- Price scraping every 15 minutes
- Cache's the top 10 most popular set prices.
- Cache's trivial ROI calculation for the most popular goods.
- Updates price db.
- ROI calculation
- Uses probability and statistical equations to calculate ROI.
- It grabs the risk and emotional value of the user via forms or NLP sentiment analysis.
- Adjust trivial ROI and create an acceptable loss limit.
- Uses Prospect and expected utility theory.
- Check reddis cache if user params are the same to avoid overhead on risk and utility re-calculation.
