In this project we tried to predict players' ELO score considering only general data about the game, without the knowledge of the precise moves made.
The data comprised mostly:
- average centipawn loss,
- the number of moves made,
- the count of blunders, mistakes and inaccuracies made by each player,
- result of the game (1-0, 0-1, 1/2-1/2),
- time control (3min, 5min etc.)
- opening played.
In our research, we tried different artificial intelligence algorithms and methods of data mining in order to find the best performing regressor.