Research pick: AI supports career choices for vocational students - "Vocational college employment training and career planning model design based on improved collaborative filtering"

A new research-led recommendation model could make vocational education more effective in guiding students towards suitable employment after graduation, according to research in the International Journal of Computational Systems Engineering. The approach has developed in response to increasing concerns in China and globally regarding youth unemployment and the efficacy of vocational training systems. As such, the algorithm reflects a more subtle approach to matching students with career paths that align better with their individual strengths and preferences.

The researchers explain that the core of their approach is based on an improved form of collaborative filtering. This is an algorithmic technique often used by streaming services and e-commerce platforms to suggest content or products based on a user’s past behaviour and preferences. The new approach overcomes various shortcomings of such systems when used in career recommendation by integrating two additional computational methods into the process: K-means clustering and the Kruskal algorithm. K-means clustering is a statistical technique that groups individuals based on shared characteristics, such as similar training choices or job applications. The Kruskal algorithm optimises how those groupings are connected and interpreted within a broader system.

The resulting model, K-means and Kruskal-enhanced Collaborative Filtering, was trained on three years of employment data from vocational school graduates. Ultimately, it was able to achieve a recommendation accuracy of more than 94 percent, significantly outperforming more conventional systems. According to the researchers, the model also excelled in standard performance metrics.

These various technical improvements represent improvements in real-world recommendations. For students, the new algorithm is more accurate in offering personalised career suggestions. And, for vocational institutions, it offers a more responsive tool to support their students in their transition from studies into the workforce. There is a third benefit and that is to employers who will gain access to candidates whose training and preferences more closely match the needs they ask for on their specific roles.

Wang, J. (2025) ‘Vocational college employment training and career planning model design based on improved collaborative filtering’, Int. J. Computational Systems Engineering, Vol. 9, No. 13, pp.1–10.