Personalized Learning Model for University-Industry Integration Through PBL Problem-Solving

Abstract

Modern educational models recommend Problem-Based Learning (PBL) and Adaptive Learning (AL) to increase the alignment of education and work organization. PBL is a technique that focuses on learning through problem-solving. However, understanding how and why PBL works is a current challenge. To advance this issue, systems that personalize learning processes are recommended. The present study developed an AL approach for PBL to align education with industry through problem-solving. The proposed approach is based on ontologies, complex multilayer networks, Bayesian networks, a recommendation system, and multi-agent architecture. The proposed tool consists of descriptive, diagnostic, predictive, and prescriptive models. This tool is available on eduCAPES with guidelines for implementing these models. Reduced PBL planning effort, increased collaboration in the development of innovation projects, enhanced educational quality, and the availability of data for Artificial Intelligence (AI) training are some benefits of this tool. To evaluate the proposed approach, nine graduate students completed the User Experience Questionnaire. The attractiveness, hedonic quality, and pragmatic quality of the models were positively perceived. Therefore, the results showed that students have a positive perception of the model's functions. The present study was limited to experiments with a small sample of students. Evaluation of the perceptions of university, educational administrators, the graduate program, the educational institution, industry, and a larger sample of students is suggested for future studies. Currently, the model is implemented with a set of existing tools. Therefore, studies to develop a platform with a user-friendly interface that integrates the proposed models are recommended.