Research on Predicting Wordle Word Attempt Counts and Word Difficulty Classification Based on Machine Learning and K-Means Clustering Techniques

Abstract

Wordle is a word-guessing mini-game that has gained tremendous popularity in recent years. As a result, there is a growing interest in analyzing Wordle's data to assist developers in problem-solving, predicting its popularity, and determining future directions. In this paper, we collected game data from Wordle users who shared their scores on Twitter between January 7, 2022, and December 31, 2022, using a Python program. We utilized machine learning and clustering techniques to develop models for predicting the number of word attempts and evaluating the game's difficulty grading. Subsequently, we conducted model testing using the word "EERIE" as an example to select the optimal model and verify its predictive accuracy. The research findings not only assist developers in enhancing user experience but also contribute to the broader field of game analytics, providing valuable insights for game design and player engagement. Ultimately, our study provides crucial data analysis support for the development of Wordle and reveals the potential and future directions of word-guessing games in the entertainment industry.