Research on the Development of Wordle Based on Multi-Model Analysis

Abstract

Wordle is becoming more and more popular around the world, so it is of great significance to study its healthy development. This paper predicts the number of Wordle 's report results and predicts the percentage of attempts for different words. Firstly, this paper establishes a prediction model of the number of ARIMA report results, and analyzes the relationship between word attributes and the percentage of players in difficult mode. This article collects data on the number of Wordle 's daily reporting results from January 7 to December 31, 2022, and predicts that the number of reporting results in 2023 will be: 10220-10637. This paper constructs three indicators to measure word attributes: the number of vowel letters, the number of affixes and the number of repeated letters. Using Pearson correlation coefficient method and AIC information criterion, according to the correlation coefficient of the three indicators, it is analyzed that there is no relationship between word attributes and the percentage of players in difficult mode. Then, this paper establishes a prediction model of the distribution of the number of attempts, and accurately predicts the percentage of attempts of EERTE words. This paper constructs the data into a lexicon and quantifies the letters. Multiple linear regression equation and multiple nonlinear regression equation were established by using the number of vowel letters, the number of affixes and the number of repeated letters corresponding to each letter in the word. The average number of guessing words and variance were fitted. It was found that the fitting effect of multiple nonlinear regression was better, was 0.805 and 0.821. Finally, the related attributes of “EERIE” were counted, and its distribution percentage was obtained. The results were 0, 1 %, 16 %, 49 %, 29 % and 3 %. The model constructed in this paper can provide some theoretical support for the good development of Wordle.