Bayesian network modelling for predicting employment tendency and major matching of vocational education students

Key findings from this study

• The study found that BEN-EM achieved higher predictive accuracy (0.75) than traditional methods while reducing mismatch rates in major-career matching.
• The authors report that the framework attained confidence scores above 0.75 and covered 65% of available career options with a Skill Alignment Score of 0.80.
• The researchers demonstrate that Bayesian network modelling captures probabilistic dependencies among academic performance, skills, interests, and labour market demand more effectively than static linear approaches.

Overview

This study introduces the Bayesian Employment and Major Matching (BEN-EM) framework to address limitations in vocational education career guidance. Traditional prediction methods rely on static surveys and linear models that lack adaptability and fail to capture probabilistic dependencies among relevant variables. The BEN-EM framework applies Bayesian network modelling to integrate multidimensional data on academic performance, skill sets, career interests, and labour market demand, enabling dynamic prediction under uncertainty.

Methods and approach

The framework leverages Bayesian network modelling to establish probabilistic relationships among student characteristics and employment outcomes. BEN-EM integrates academic performance metrics, skill assessments, career interest profiles, and labour market data within a unified probabilistic structure. The approach enables vocational institutions, policymakers, and career counsellors to identify optimal major-career pathways and support evidence-based decision-making.

Results

Experimental validation demonstrated that BEN-EM achieved a predictive accuracy of 0.75, representing higher performance than conventional methods. The framework exhibited greater adaptability with scores exceeding 2.5 and produced lower mismatch rates between student capabilities and employment requirements. BEN-EM attained confidence scores above 0.75 and provided coverage of 65% of available career options, with a Skill Alignment Score of 0.80 and decision support effectiveness scores of 0.75–0.80.

Implications

The framework enables vocational education institutions to enhance career readiness and reduce skill-job mismatches through probabilistic prediction models. BEN-EM supports evidence-based recommendations that improve alignment between student competencies and employer requirements, potentially strengthening employment outcomes for vocational education graduates. The dynamic integration of multidimensional data facilitates responsive decision-making by career counsellors and policymakers in evolving labour markets.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.