Traditional employment selection methods can no longer meet the individualized and diversified career needs, especially in the context of the rapid development of the Internet and big data, college students are faced with numerous choices and complex decision-making factors in employment and entrepreneurship selection. This paper proposes a multi-objective dynamic path planning-based career choice model for college students’ employment and entrepreneurship (RJMGP), which aims to optimize the career choice paths of college students by comprehensively considering their personal preferences and the demands of the global job market. The study first constructs a reciprocal recommendation model based on global preferences, and formulates a recommendation quality assessment function by introducing a balance between global optimality and personal preferences. In order to improve the computational accuracy and convergence speed, the improved particle swarm optimization algorithm (SAMOPSO) is adopted, and several improvements are made to the traditional particle swarm algorithm, such as the introduction of Logistic chaotic mapping initialization and adaptive learning factor strategy. The experimental results show that SAMOPSO performs well in several test functions, especially when dealing with multi-objective optimization problems, and exhibits better performance than the other three algorithms. In terms of specific application, by making career recommendations for 100 students, the SAMOPSO model achieves a significant improvement in career recommendation satisfaction compared to the traditional method, with an average satisfaction of 8.7, compared to 7.1 for the traditional recommendation method. The experimental results validate the effectiveness and feasibility of the proposed model.
