应用组合模型的高校毕业生就业率预测研究  

Study on the Prediction of Employment Rate of College Graduates Based on the Combination Model

作　　者：徐永慧 XU Yonghui(Ningxia Center for Educational Technology, Yinchuan 750004, China)

机构地区：[1]宁夏电化教育中心,宁夏银川750004

出　　处：《微型电脑应用》2021年第10期154-156,共3页Microcomputer Applications

摘　　要：当前高校毕业生就业受经济环境、就业市场和自身发展情况等因素的影响,具有随机性和周期性的变化特点,因此高校毕业生就业率是一种复杂的非线性系统,单一模型无法对该变化特点进行全面描述。为了降低预测偏差、增强结果可信度,获得更优的高校毕业生就业率预测结果,提出了组合模型的高校毕业生就业率预测方法。采用小波分析对高校毕业生就业率数据序列进行多尺度分解,得到高校毕业生就业率的子序列,利用灰色模型确定累加序列参数,通过BP神经网络得到误差反馈值。仿真测试表明,组合模型可以有效提高就业率预测能力、强化高校毕业生就业率预测精度为95.94%左右,优势十分明显。The current employment of college graduates is affected by factors such as the economic environment,job market,and their own development.It has the characteristics of random and periodic changes.Therefore,the employment rate of college graduates is a complex nonlinear system.The characteristics of the changes are fully described.In order to reduce the prediction bias,enhance the credibility of the results,and obtain a better prediction result of the employment rate of college graduates,a combination model of college graduate employment rate prediction method is proposed.The wavelet analysis is used to decompose the data sequence of the employment rate of college graduates in multiple scales,and the sub-sequence of the employment rate of college graduates is obtained.The gray model is used to determine the cumulative sequence parameters,and the error feedback value is obtained through the BP neural network.The simulation test shows that the combined model can effectively improve the predictive ability of employment rate and strengthen the accuracy of predicting the employment rate of college graduates,with an accuracy of about 95.94%,which has obvious advantages.

关 键 词：毕业生就业率 小波分析 组合模型 灰色模型 BP神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] G473.8[自动化与计算机技术—计算机科学与技术]