A Perspective on the Current Situation of Children’s Education and Teachers’ Moral Cultivation Combined with Decision Tree Algorithm

Abstract

The enhancement of children’s education quality necessitates a high level of teacher moral literacy. This paper employs data mining techniques to elaborate on the Classification and Regression Tree (CART) model within logistic regression and decision tree algorithms. Data pertinent to elementary school education were collected via questionnaires, followed by data cleaning and segmentation. The parameters of the CART model were optimized using a Bayesian algorithm, and its efficacy was validated through a dataset to support the exploration of factors influencing the developmental level of children’s education. Additionally, data regarding teachers’ moral literacy were gathered and analyzed using binary logistic regression within the logistic regression framework. The analysis revealed that the accuracy of the CART decision tree model reached 90%, with the error converging to zero at 50 trees. Key determinants such as willpower and communication abilities in children’s education were influenced by exercise duration, time spent in interest classes, gender, and parents’ educational levels. The mean score for teachers’ moral literacy was 4.349±0.465. Notably, teachers with low salary satisfaction exhibited a moral literacy level that was only 0.517 times that of their counterparts with high salary satisfaction. Establishing a scientifically sound and reasonable salary management system is crucial. Such a system can ignite teachers’ enthusiasm for teaching, substantially enhancing their moral literacy, which in turn positively impacts the level of children’s educational development.

Keywords

• data mining
• decision tree
• cart model
• logistic regression
• bayesian algorithm
• child education
• 97p10