-
学术期刊
-
学术著作
-
精品图书
-
投稿须知
Technological Innovation and Science Management(TISM) June 2025, Volume 1, Issue 1
Predicting Wordle Results
Yikun Chen,Tiantian Wang
Zhengzhou Technology and Business University,Zhengzhou,Henan,451400
Summary: Wordle is the latest popular crossword puzzle, made up of simple tables with six rows and five columns. With the spread of wordle on the Internet, the number of users is growing rapidly. Therefore, this paper attempts to establish a mathematical model to solve the fol-lowing problems.
For the first part: In order to explain the change of the number of reported results, an ARIMA time series prediction model is established, and the historical data are used to simulate the number of results reported on March 1, 2023. The prediction interval is [9968.541,10995.79]. For the question of whether the word attribute will affect the score percentage in the difficult mode, we first classify the words according to the word attribute, and then establish a linear regression model. finally, the fitting degree of the regression curve is 0.576, which is quite different from 1. In the end, it is concluded that any attribute of the word EERIE will not affect the percentage in the Hard Mode.
For the second part: In order to predict the correlation percentage of the future date (1, 2, 3, 4, 5) the five letters involved in the words are encoded, and the BP neural network
prediction model is established. The error analysis is carried out according to the results of
each training and the expected results, and then the weights and thresholds are modified until the model is consistent with the expected results. Finally, after the expression between the
input data and the output data is obtained through the model, the correlation percentage of (1,2,3,4,5,6,X) predicting the future date word EERIE is 0.7,8.6,24.2,31.3,23.4,11.2,1.6.
For the third part: First of all, the dimension of the data in the attachment is reduced, and the correlation analysis of the data shows that KMO=0.847 > 0.5. therefore, it is suitable for principal component analysis, from which the principal component analysis model is
established. finally, the word is analyzed and defined, which is divided into simple and difficult attributes, and finally the word EERIE is defined as simple.
For the fourth part: We make a box chart comparison of the relevant percentages (1, 2, 3, 4, 5, 6, X) and find that the data set is hillside.
Keywords: ARIMA time series prediction model, linear regression model, BP neural network prediction model, principal component analysis model
