Wordle character prediction model based on linear XGBoost

This paper focuses on predicting four questions posed by the New York Times about Wordle. We use a least-square linear model and a nonlinear XGBoost model to create a regression model that predicts the percentage correlation for each of the future days (1, 2, 3, 4, 5, 6, X). Then, we combine these two models to form a linear XGBoost model, which combines the advantages of linear and nonlinear models and can greatly enhance the accuracy of the prediction's fit.