With the rapid development of cloud computing technology and big data analysis, job search and employment services for college students are gradually evolving in the direction of intelligence and precision. This paper proposes a personalized job recommendation model that integrates dynamic behavior modeling and improved collaborative filtering algorithm. An elastic resource scheduling framework based on cloud computing is constructed, and the response efficiency of the system is improved through inertia weight optimization and task allocation strategy. A behavioral-interest model is established to solve the problem of dynamic updating of user preferences. Introduce time decay factor to improve collaborative filtering algorithm and enhance the timeliness and accuracy of recommendation system. Based on the data of finance and economics graduates from University B, the average accuracy of the improved algorithm and the original algorithm are 0.47 and 0.67 respectively, and the average recall is 0.51 and 0.65 respectively, which proves that the improved algorithm in this paper is able to effectively match the interests of job seekers with the characteristics of the jobs, and it has the value of application in the college students’ career planning.
