Exploring the Path of Cultivating and Managing Graduate Students' Innovative Ability Based on Decision Tree Modeling

Abstract

Graduate students' innovation ability is the main driving force to promote academic research and scientific and technological progress, but at present there are still many difficulties and obstacles in the cultivation and management path. In this paper, we firstly collected the relevant information of the students in the School of Information Science and Technology of X University of Science and Technology, and constructed the evaluation index system of graduate students' innovation ability from the three aspects of comprehensive innovation thinking ability, innovation quality characteristics and innovation practice ability. Based on the random forest to predict the innovation ability of graduate students, then quantitatively analyze the characteristic importance of the factors influencing the innovation ability of graduate students under different grades by SHAP model. Then the partial dependency diagram was used to further reveal the relationship existing between the influencing factors and graduate students' innovation ability. The prediction accuracy, precision, recall and F1 value of the model are all above 80%, and the importance of the influencing factors varies with different grades of graduate student innovation ability. Finally, this paper proposes a path for the cultivation and management of graduate students' innovation ability at three levels, namely, innovation of interdisciplinary knowledge system, adoption of interactive teaching methods and enrichment of practical programs for innovation ability cultivation, in order to promote the high-quality development of graduate education.