Designing a precision career-guidance model based on student psychological profiling in higher education
Introduction In the rapidly evolving landscape of higher education, the demand for personalized and data driven career guidance has become increasingly critical. Traditional career counseling methods often rely on standardized assessments and static recommendations, which fail to capture the diversity of student characteristics, learning preferences, and evolving aspirations. Methods To address this challenge, this paper proposes the Psychological Profiling Driven Career Guidance Model, abbreviated as PPDCGM, for precision career guidance in higher education. The framework integrates psychologically informed profiling with machine learning techniques to generate personalized career recommendations. The Adaptive Career Pathway Strategy, abbreviated as ACPS, dynamically refines recommendations according to changes in learner states, skill development, and external career conditions. For reproducible evaluation, the framework is implemented on public career trajectory benchmarks and assessed through next career prediction as a practical proxy for precision career guidance. Results and discussion Experimental results show consistent improvements over competitive baselines. On KARRIEREWEGE, the proposed method achieves an MRR of 0.398 and a Recall@10 of 0.667, surpassing the strongest baseline MiniLM by 0.024 in MRR and 0.029 in Recall@10. On DECORTE, it achieves an MRR of 0.371 and a Recall@10 of 0.641, with gains of 0.024 and 0.027 over MiniLM, respectively. These findings demonstrate the effectiveness, robustness, and practical value of the proposed framework for intelligent educational recommendation.