The problem of incorrect analysis and forecasting is a significant worry in the students, despite the importance of employment prediction. When considering vocational student employment analysis, traditional deep learning is ineffective and unable to provide satisfactory results. Given this, the study both analyzes and offers a method for analyzing and predicting career prospects. To minimize interference elements in employment analysis and forecasting, we first use information theory to identify the components that have an impact, and then we split the indicators based on the needs of the two processes. Subsequently, the decision tree classification algorithm’s employment analysis and prediction scheme are developed using information theory, and the outcomes of this process are thoroughly examined. When compared to traditional deep learning, the decision tree classification method outperforms it in MATLAB simulations when it comes to time spent analyzing and predicting influencing factors in the job market, as well as accuracy in employment analyses and predictions.