“互联网+”视角下基于WOA-BPNN高职院校大学生就业评价

Employment evaluation of college students in higher vocational colleges based on WOA-BPNN from the perspective of"Internet+"

为提高高职院校大学生就业质量评价的精度,针对反向传播神经网络BPNN模型性能受其参数选择影响,选择鲸鱼优化算法WOA对BPNN模型进行优化的WOA-BPNN模型高职院校大学生就业质量评价模型。首先,从就业状况、就业结构、就业质量以及满意度等4个四个方面,建立了一套基于层次分析法的高职院校大学生就业质量评价指标体系;其次,针对初始权值和阈值对BPNN模型性能影响的问题,为避免陷入局部最优,运用WOA算法对BPNN模型的初始参数的权重值以及阈值进行优化,建立WOA-BPNN的高职院校大学生就业质量评价模型。与PSO-BPNN和BPNN对比可知,WOA-BPNN进行高职院校大学生就业质量评价具有更高的分类准确率、特异性以及灵敏度。

In order to improve the accuracy of the evaluation of the employment quality of college students in higher vocational colleges,the performance of Back Propagation Neural Network(BPNN)model is affected by its parameter selection,and a Whale Optimization Algorithm(WOA)is proposed.Optimize the BPNN(WOA-BPNN)evaluation model of employment quality of college students in higher vocational colleges.First,from the four aspects of employment status,employment structure,employment quality and satisfaction,a set of evaluation index system for the employment quality of college students in higher vocational colleges based on the analytic hierarchy process was established;Secondly,aiming at the influence of the selection of initial weights and thresholds on BPNN performance,to avoid falling into local optimal,the WOA algorithm is used to optimize the initial weights and thresholds of the BPNN model,and a WOA-BPNN employment quality evaluation model for college students in vocational colleges is established.Compared with PSO-BPNN and BPNN,it can be seen that WOA-BPNN has higher classification accuracy,specificity and sensitivity in evaluating the employment quality of college students in higher vocational colleges.