Current talent introduction strategies are mainly based on staff arrangement, school discipline construction and so on, which depend on experience actually. However, this kind of empirical approach, lacking of scientific basis, usually causes problems in applications such as uneven scientific research level. In this paper, we intend to use data mining to analyze talent information of teachers in University of Finance and Economics, China from 2011 to 2017, and then to predict their capabilities in obtaining National Foundation of China. In a word, this paper aims to provide decision support for universities’ talent introduction strategies. After data cleaning and feature engineering, Apriori algorithm is applied to mine the association rules and find key factors that are closely related to teachers' acquisition of National Science Foundation of China. Then we make predictions with four kinds of models, including Logistic Regression Model, Decision Tree Model, Artificial Neural Network Model and Support Vector Machine Model. In the end, in order to get a more accurate model, Logistic Regression Model which has the highest accuracy of prediction is used to do stepwise regression.

talent introduction strategies, apriori algorithm, prediction model, R language.