A Knowledge-Driven Approach to AI-Based Personalized Test Paper Creation in Programming Education

Abstract

Computer programming education plays an important role in modern society, however, students' knowledge base, learning style, and ability level vary widely. Personalized teaching has become a key issue in education reform. As the existing question bank system mainly labels the difficulty coefficient of the questions by manual means, and cannot objectively and accurately provide personalized learning paths for different students. This paper puts forward two parameters of “students' ability value” and “difficulty value of test question”, constructs the corresponding relationship table between students' ability value and difficulty value of test question, establishes the mathematical model, and generates the test paper that meets students' learning ability. Compared with other algorithms, the algorithm designed in this paper has a faster convergence speed and the success rate of paper generation. Through practical application and questionnaires, it is evident that the personalized question bank designed in this paper can effectively help students better grasp programming knowledge.