Report on the Analysis and Prediction of Wordle Data Based on the SIR Model

Abstract

Wordle is a New York Times crossword puzzle that has become very popular recently. In this paper, a SIR infectious disease model was developed to explain the reasons for the variation in the number of daily reported results and to predict the number of future reports. By building the SIR contagion model, this paper explains that the main reason for the change in the number of daily reports over time is Twitter publicity, and predicts that the number of reported results on March 1, 2023 will be approximately 9411. Further, by analyzing the correlation and feature importance ratings from decision tree regression, we conclude that words with common letters and high frequency are more likely to be guessed by players with fewer guesses, whereas words with repeated letters were less favorable. Also, the presence or absence of common letters in words had the most significant effect on the difficulty of the game.