There is a significant trend of diversification in the development of vocational education students, and the traditional prediction methods have insufficient accuracy to meet the needs of personalized learning. The support vector machine algorithm has a unique advantage in the prediction of small-sample high-dimensional data, which provides a new idea to solve this problem. In this study, we constructed a support vector machine prediction model based on the optimization of differential evolution algorithm, and explored the methods of predicting the development trend of vocational education students and personalized learning path planning. The study adopts SVM algorithm to deal with nonlinear high-dimensional data, introduces differential evolution algorithm to solve the problem of SVM parameter selection, and realizes regression prediction through insensitive loss function. The experiment selects 500 sample data for training and 60 sample data for testing, and the student development trend prediction model reaches the accuracy requirement of absolute error limit value 1 after 115 steps of training. An empirical study was conducted on 804 valid student data of a vocational college, and the overall accuracy of the prediction model reached 98.13%, of which the highest accuracy was 94.3% for the item of “employment” and 91.6% for the item of “higher education”. The results of this study show that the student development prediction model based on support vector machines can effectively predict the development direction of vocational education students and provide data support for vocational colleges to develop personalized learning paths. The study further proposes a personalized learning path design framework, including four key aspects: data collection and management, learning portrait and demand orientation, learning goal and path planning, and learning resources and support services construction, to provide precise guidance for vocational education students’ development.
