AIversonBets

Title page
Prediction form
Player options
Test Result

Inspiration

The inspiration for AIverson Bets came from a desire to combine advanced machine learning techniques with sports analytics. By leveraging the vast amount of NBA data available, we aimed to create a tool that could provide accurate predictions on player performance. This not only serves sports enthusiasts and bettors but also showcases the potential of AI in sports analytics. The inspiration behind the name comes from the NBA legend Allen Iverson who also goes by AI.

What it does

AIverson Bets predicts the points a specific NBA player will score against a particular team based on historical game data. Users can select a player, an opponent team, and the number of recent games to be considered for the prediction. The app then uses a trained machine learning model to forecast the player's performance and displays the prediction.

How we built it

Data Collection: Used the NBA API to fetch historical game data for the selected players. Data Preprocessing: Cleaned and organized the data, focusing on points scored and matchup details. Model Training: Trained an XGBoost regression model using the preprocessed data. Web Application: Developed a Flask-based web application to provide a user-friendly interface for making predictions. The front-end was styled using CSS. Deployment: Hosted the application locally for development and testing purposes.

Challenges we ran into

Data Fetching: Handling API rate limits and timeouts while fetching large amounts of historical data. Feature Engineering: Ensuring that the features used for model training were relevant and properly formatted. Model Accuracy: Achieving high prediction accuracy required tuning the model and experimenting with different machine learning techniques. UI/UX: Creating a seamless and intuitive user experience.

Accomplishments that we're proud of

Model Performance: Successfully training a model that provides reasonably accurate predictions based on historical data. User Interface: Developing a clean, responsive web application that allows users to easily interact with the model and visualize results. Integration: Efficiently integrating various technologies (Flask, CSS) to create a cohesive application.

What we learned

API Handling: Techniques for efficiently fetching and managing large datasets from APIs, including handling timeouts and retries. Machine Learning: Improved understanding of feature engineering, model training, and performance evaluation. Web Development: Enhanced skills in building full-stack web applications, particularly with Flask and front-end integrations.

What's next for AIversonBets

Model Improvements: Continuously improve the prediction model by incorporating more features and experimenting with different algorithms. Expanded Metrics: Add predictions for other performance metrics such as rebounds, assists, and player efficiency ratings. Deployment: Deploy the application on a cloud platform for broader accessibility and scalability.