基于RBF神经网络的科研绩效评价建模研究

发布时间：2018-06-27 08:35

  本文选题：绩效评价 + 粒子群优化　； 参考：《江苏科技大学学报(自然科学版)》2017年04期

【摘要】：客观、公正、准确的科研绩效评价是调动和提高高校及科研机构科研人员工作积极性和科技创新能力的重要措施.文中提出了一种基于RBF神经网络的科研绩效精细评价模型,以归一化后的科研指标数据乘以相应权系数作为网络输入,利用优、良、中、及格和不及格5级评价作为输出,采用粒子群优化算法通过交叉验证对RBF网络结构参数进行了优化.通过RBF网络结构和输入输出特性分析发现,训练后的RBF网络权值与5级评价结果高度相关,并较5级评价结果更能精细区别科研绩效差异.该权值可直接用来进行科研绩效精细评价.文中推广了RBF网络在科研绩效评价中的应用,并为进行类似评价或评估工作提供了一种新思路.
[Abstract]:Objective, fair and accurate evaluation of scientific research performance is an important measure to arouse and improve the enthusiasm of scientific research personnel and the ability of scientific and technological innovation in universities and scientific research institutions. In this paper, a fine evaluation model of scientific research performance based on RBF neural network is proposed. The normalized scientific research index data multiplied by the corresponding weight coefficient are taken as the input of the network, and the five grades of excellent, good, moderate, passing and failing are used as the output. Particle swarm optimization (PSO) algorithm is used to optimize the parameters of RBF network structure. Through the analysis of RBF network structure and input and output characteristics, it is found that the weight value of RBF network after training is highly correlated with the evaluation results of level 5, and it can distinguish the difference of scientific research performance more carefully than the results of evaluation of level 5. The weight value can be directly used for fine evaluation of scientific research performance. In this paper, the application of RBF neural network in evaluation of scientific research performance is extended, and a new idea is provided for similar evaluation or evaluation.
【作者单位】： 江苏科技大学经济管理学院;
【基金】：国家自然科学基金青年项目(71303096) 2016年校研究生教育教学改革研究与实践项目(YJG2016Y_14)
【分类号】：G311;TP183
，

本文编号：2073328