Predicting Graduates’ Career Paths Using Neural Networks and Gaussian Mixture Models

To promote the contribution of university graduates to the socio-economic development, an increasing number of universities have begun to focus on the employment prospects of their students after graduation. Meanwhile, the changes in various professions in society have provided important directions for talent cultivation in universities. Therefore, universities have generally improved their teaching methods, evaluation indicators, and provided better employment guidance services. However, in addition to collecting data on graduates, universities should also plan for data on the employment prospects of students who are preparing to graduate, to provide better guidance and services. This paper proposes the use of a certain university’s employment data set for graduates to construct feedforward neural network and Gaussian mixture models to predict the employment prospects of students. Finally, the model’s prediction results are evaluated using AUC as the evaluation metric to verify the performance of the model in predicting the employment prospects of graduates.