Traditional human resource evaluation methods rely too much on people’s subjective judgment and experience. In order to overcome this defect, a talent management model based on decision tree and correlation analysis is proposed. The scheme uses the concept of decision tree, classifier and correlation analysis to preprocess the primary data. Then Apriori association rules and C4.5 algorithm are adopted to complete the combination prediction of the two, which reduces the efficiency of the algorithm to a certain extent when searching for approximate high reliability rules, and makes the improved decision tree algorithm have good scalability. Finally, the model is tested and analyzed through the actual data in human resource management in universities. The results show that the scheme can improve the prediction accuracy of the algorithm and provide a theoretical basis for scientific decision-making of leaders and functional departments.