Research on word difficulty classification based on machine learning

To classify the solution words according to difficulty and determine the attributes of the given words, a model is developed. The model is used to classify the word "EERIE", and the accuracy of the classification is discussed. In this study, the average percentage distribution of unequal weights is used to quantify the number of words divided into three levels: easy, medium, and difficult, and then tagged. In this study, the radical support vector machine model with radial basis function kernel is used, the accuracy is 0.6556 and the F1 value is 0.6634. In addition, the classification result of EERIE is difficult. This study also found some interesting observations from the data set, one of which discussed the difference between human thinking and machine learning.