Interdisciplinary Competence Prediction for Engineering College Students Based on Decision Tree

With the increasing complexity of engineering problems, it is difficult for single-discipline knowledge to solve complex engineering problems, and the cultivation of interdisciplinary ability has become the main goal of engineering education. However, the process of interdisciplinary competence development is affected by many factors, such as student factors, teacher factors, school factors, and artificial intelligence factors. Firstly, we collected data on the influencing factors of interdisciplinary competence, and secondly, we performed data processing and explored the importance of each influencing factor on interdisciplinary competence using the classification and regression trees algorithm. The prediction model and 7 decision rules were obtained after optimizing the decision tree using the cost complexity pruning algorithm, the model has an accuracy rate of 75.1%. Finally, we analyze the experimental results and conclude that teachers' teaching style is the main factor affecting interdisciplinary competence, followed by the frequency of using smart tools, students' motivation to be interdisciplinary, the availability of school facilities, teachers' interdisciplinary background, and school curriculum as important influencing factors. The results of this study will provide applied guidance for the development of interdisciplinary competence among engineering college students.