XGBoost-Based Analysis of a Multi-disciplinary Talent Training System for Enhancing International Competence

In the context of globalization and national development strategies, the demand for globally competent talent is with an increased tendency. To this end, this study proposes a data-driven interdisciplinary talent training model. The model is based on the principles of integration, innovation and internationalization. The goal of this model is tantamount to training graduate students. These students should not only have a solid academic foundation, being fluent in foreign languages and having intercultural communicative competence is one of the key criteria. It supports the integrated development of professional knowledge, international vision and cultural adaptability by merging the content of curricula both inside and outside of class with each other and organizing more participatory, interactive seminars and workshops. Using online survey data as a test of the model, a survey that contains 354 valid results concerning the current status of interdisciplinary education in global business programs in Asia is carried out, full capacity has been reached. By using the XGBoost machine learning algorithm to construct a classification model, it has been able to predict students' level of international capability (low, medium, and high). The characteristics which have the greatest effect on students' capability levels have also been identified. The key evaluation tools consist of confusion matrix, feature importance map, and ability level distribution map, all used to evaluate the performance. The model's genesis and functionality is highlighted and explained through this analysis, which pinpoints student engagement with overseas studies, students’ reflection on real-life applications, and the readiness of students to get involved in interdisciplinary work as the greatest contributors to a person's high level of global competence. The study results indicate that the proposed model has useful functions as well as strong rationality from a theory perspective; additionally, the results provide actionable input for Chinese university administrators to help initiate reform measures to their graduate programs so that Chinese universities will have prepared for future global demands.