Machine Learning Based Preschool Education Quality Assessment System

Abstract

Preschool education (PE) is the initial stage of life education, and early childhood is an unrepeatable process. PE has the same importance as other education stages because of the significant impact it can have on later childhood development. Furthermore, from the perspective of educational equity theory, every child has the right to receive PE, the right to obtain the same high-quality educational resources, and the right to fair final results. Therefore, the research on the quality of PE has theoretical value and practical significance. In order to strengthen the quality of PE, this paper designs a PE quality assessment system to evaluate teachers’ teaching achievements. In this regard, the performance of each functional module in the system is tested, and the test results show that the module access is successful at more than 97%, indicating that the system meets the operating requirements. This paper uses the characteristics of the KNN algorithm classification in the machine learning algorithm to classify the teaching quality (TQ) of 7 pre-school teachers, and obtains the membership degrees of teachers in the four categories of grades, indicating that the KNN algorithm is more suitable for the classification of TQ assessment results than the general classification algorithm.