Research On Lexical Data Set Analysis Based on Decision Tree Model

Abstract

In order to predict the future number of users and development prospects of "World" game, this paper analyzes the terminology used in game data sets. Firstly, descriptive statistical analysis was used to analyze the existing data, and ARIMA model was used to estimate the data during the forecast period, so as to obtain the interval estimation and point estimation of the number of results. Secondly, EXCEL is used to calculate the percentage of each word, and analysis of variance model is used to get the attribute influence between words. Thirdly, the training set and test set were analyzed through machine learning, and the mapping model was established. "EERIE" was input as the word vector to obtain the prediction results. Finally, the relationship between the decision tree model and the actual expected break time is established, and the difficulty of EERIE is evaluated. The results show that with the word vector of EERIE as input, when the model accuracy is 55.56%, the prediction result is (1,2,3,4,5,6,X)-(6,12,21,33,25,3,0). The prediction difficulty was medium, and the accuracy of the model was 80%.