League of legend recomendation model is a data science project including winrate prediction that we engieering the feature from
Kinkade, N. and Yul, K., 2015. DOTA 2 Win Prediction. online. Available at: http://jmcauley.ucsd.edu/cse258/projects/fa15/018.pdf and make the recomendation model based on unsupervised learning model (e.g. K-nearest neightbor) and association rule based (Apriori algorithm)
Use the package manager pip to install required library.
pip install requirements.txt
flask run
For our experimnet the accuracy for winrate prediction model, we improve the accuracy from 0.55 to 0.67 (using decision tree) and test accuracy from 0.62 to 0.72 (using gradient boosting) by using champion synegy/countering on blue side and red side.
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Please make sure to update tests as appropriate.