Fuzzy cognitive maps (FCMs) are a powerful tool for system modeling, which can be used for static and dynamic analysis. However, traditional FCMs are usually learned by gradient-based methods, and the adopted sigmoid nonlinear activation function frequently causes gradient saturation. These two shortcomings set a limit on the modeling accuracy. To overcome those problems, we propose in this paper a new FCM with two improvements. First, the rectified linear unit (ReLu) activation function is adopted to replace the sigmoid function. Second, a newly proposed quasi-oppositional bare bone imperialist competition algorithm (QBBICA) is used to learn the FCM. The improved FCM is used to predict the employment rate of graduates from Liren College, Yanshan University. Experimental results show that the improved FCM is effective in employment rate prediction.

Prediction of the employment rate of college graduates is very important not only for education department to make suitable decision but also for university to adjust its teaching plan. However, it is a difficult task to accurately predict the employment rate of college graduates because it is affected by many factors, making the original data fluctuant. In this paper, a simple and effective prediction approach is proposed to predict employment rate of college graduates using GM (1,1). To attenuate effect of data fluctuation and improve prediction accuracy of GM (1,1) model, the geometric average weakening buffer operator (GAWBO) is first applied to original employment rate data, and then GM (1,1) model is used to predict employment rate. The prediction experiment of Yanshan university employment rate from 2010 to 2019 is performed using the proposed method. Results show that the proposed method can obtain more accurate prediction result than GM (1,1). © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC.