Using Ensemble Machine Learning Algorithms to Predict a Scrabble Player's Rating

The rapid development of machine learning technology has increased its popularity, especially in addressing regression problems. Machine Learning algorithms play a crucial role in discovering hidden patterns in data, enabling handling complex challenges and making reliable decisions at large scale. One such challenge is determining the skill level of a Scrabble player before engaging in a game. To tackle this, we have constructed a unique dataset comprising thirty features derived from the publicly available dataset for Scrabble gameplay on Woogles.io, which was made accessible through a Kaggle competition. Our objective was to predict the Scrabble player rating using a variety of ensemble machine-learning techniques, including Xgboost, Adaboost, Random Forest, Bagging, and Gradient Boosting. According to our findings, Random Forest exhibited exceptional performance when evaluated externally on the Kaggle platform, with the lowest Root Mean Square Error value of 108.19.