An Accurate Financially-Challenged College Student Identification Model Based on Rough Set-BP Neural Networks and Its Application

The previous methods for identification of financially-challenged college students are subjective and liable to influence of human factors. Against the backdrop of smart campus and with data collected based on college students' campus visas, the rough set attribute reduction theory and the neural network technology were combined to establish a new model for accurate identification of financially-challenged college students. First, with campus big data, data on the students' consumption behaviors and learning attitude were analyzed through visualized discrete analysis methods; after that, the rough-set-based decision-making software was used to reduce the attributes and identify the core influencing factors; then, 90 sets of reduced data were selected randomly and input to the BP neural network for training, and thereby an accurate identification model was established; at last, ten sets of reduced data were selected randomly for verification of the proposed model. The results show that the proposed model effectively reduced the training time, achieved high prediction accuracy and showed strong adaptability.