Research on Multi-Objective Optimization Recommendation Algorithms for Work-Study Platforms

Abstract

With the rapid advancement of digital technologies, personalized learning platforms have become increasingly important in vocational education. This study, based on the digital platform of Xiamen Nanyang Vocational College, leverages dynamic knowledge graphs, multi-objective optimization, and adaptive algorithms to provide students with personalized learning resources and work-study position recommendations. By analyzing students' learning behaviors—such as course completion rates, exam scores, and interaction logs—the platform constructs a dynamic knowledge graph that accurately reflects students' mastery of specific topics. A multi-objective optimization framework is designed to balance educational relevance, user preferences, and learning outcomes, incorporating objectives such as Knowledge Coverage (KC), Interest Matching (IM), and Goal Achievement (GA). The Adaptive Multi-Objective Particle Swarm Optimization (AMOPSO) algorithm is developed to generate Pareto-optimal recommendation sets, ensuring that students receive diverse, relevant, and goal-oriented suggestions. The proposed methodology not only enhances the quality of personalized recommendations but also improves student engagement, academic performance, and career readiness. This research contributes to the development of intelligent, adaptive, and student-centered educational systems, offering valuable insights for educators, policymakers, and industry stakeholders.