基于BP神经网络的企业管理人才评价研究
发布时间：2018-09-05 07:02
【摘要】：企业管理人才评价是企业人力资源管理实践的重点,管理人才的科学评价和选拔是企业获取核心竞争能力的有效途径,也是企业实现人尽其才、优化人才资源配置的前提。目前对于企业管理人才的评价,存在较大的主观性,主要表现在:评价指标的权重设置过多依靠专家经验,无法结合企业实际,导致评价指标体系难以推广应用;管理人才评价过程中,人才自身参与度低,单纯依靠上级领导的主观评价,使得评价结果的认可度低,容易打击管理人才的积极性。随着人才评价研究的不断推进和深入,各种各样的人才评价方法都被运用到人才综合评价过程中,但是人才评价如何与企业的用人实际相结合,符合企业的用人倾向,将企业人才评价工作与企业的文化和发展战略相结合,开发和构建适合企业自身特点的管理人才评价体系,是本研究要解决的重点问题。本研究通过对国内外关于人才、管理人才、人才评价等方面的相关理论和目前研究中运用的人才评价方法进行梳理和总结,结合文献分析,根据频度统计的结果,筛选出17个影响企业管理人才能力水平的评价指标,借鉴前人关于人才、管理人才理论、素质结构模型的研究成果,建立了一个含有4个一级指标、17个二级指标的企业管理人才评价指标体系。由于人才评价的特殊性,为了更好的利用人才之间的差异性特征,选用熵值法来确定评价指标的权重,通过BP神经网络建模得到评价对象的综合得分。然后构建BP神经网络,通过网络学习和测试,建立企业管理人才评价体系。BP神经网络学习的重点在于选取学习样本集和训练样本集,为了实现研究目的,网络的学习样本集分为两部分:一部分为前面研究所得到的熵值法计算综合得分的样本数据,这一部分数据代表专家经验;另一部分是反映企业用人实际和用人倾向的样本数据,选择企业管理人才的薪资水平作为综合评价得分(薪资水平是企业在多年人才选拔和人才评价的过程中的经验总结,是不同企业不同用人倾向的实际反映,是人才能力与岗位匹配的实践成果)。将两个部分的学习样本组合,通过训练以后的网络就能既反映专家经验,又体现企业在人才选拔和任用过程中的倾向。收集的企业用人实际的那部分学习样本又可以作为测试样本集,用来测试神经网络的精度。通过测试的神经网络,测试样本评价误差都在5%以内,说明建立的神经网络模型具有良好的评价能力,可以用于后续企业管理人才评价。本文利用BP神经网络实现对企业管理人才进行评价,不仅是对现有的企业管理人才评价方法的补充,也能避免因主观因素造成的评价偏差,同时能够建立一个既能借鉴专家经验,又符合企业用人实际的企业管理人才评价系统,更好的服务于企业的人才评价实践。本研究建立的企业管理人才评价系统,希望可以对企业管理人才的选拔和聘用等工作提供一定的借鉴。
[Abstract]:Enterprise management personnel evaluation is the focus of human resource management practice. Scientific evaluation and selection of management personnel is an effective way for enterprises to acquire core competitiveness. It is also the premise for enterprises to make full use of their talents and optimize the allocation of human resources. It is difficult to popularize and apply the evaluation index system because the weight of the evaluation index is too much depended on the expert experience and can not be combined with the actual situation of the enterprise. With the continuous advance and deepening of price research, various methods of talent evaluation have been applied to the process of comprehensive evaluation of talents. However, how to combine talent evaluation with the actual employment of enterprises, accord with the employing tendency of enterprises, combine enterprise talent evaluation with enterprise culture and development strategy, develop and construct suitable for enterprises themselves This research combs and summarizes the relevant theories of talent, management talent, talent evaluation at home and abroad, and the talent evaluation methods used in the current research. Combined with literature analysis, 17 influential enterprises are selected according to the results of frequency statistics. Based on the previous research results on talent, management talent theory and quality structure model, an evaluation index system of enterprise management talent is established, which contains four first-level indicators and 17 second-level indicators. Entropy method is used to determine the weight of the evaluation index, and the comprehensive score of the evaluation object is obtained by BP neural network modeling. Then BP neural network is constructed to establish the evaluation system of enterprise management personnel through network learning and testing. The learning sample set of the network is divided into two parts: one is the sample data obtained by the entropy method, which represents the experience of experts; the other is the sample data reflecting the actual employment and Employment Tendency of enterprises, and the salary level of enterprise management personnel is selected as the comprehensive evaluation score (salary water). Ping is a summary of the experience of enterprises in the process of selecting and evaluating talents for many years. It is a reflection of different employing tendencies of different enterprises and a practical result of matching talents and posts. The actual part of the learning sample collected by the enterprise can be used as a test sample set to test the accuracy of the neural network. Through the test of the neural network, the evaluation error of the test sample is less than 5%. This shows that the neural network model has good evaluation ability and can be used in the subsequent enterprise management. In this paper, BP neural network is used to evaluate enterprise management talents, which is not only a supplement to the existing evaluation methods of enterprise management talents, but also can avoid the evaluation deviation caused by subjective factors. This research establishes the enterprise management talent evaluation system, hoping to provide some reference for the selection and employment of enterprise management personnel.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP183;F272.92
本文编号：2223538
[Abstract]:Enterprise management personnel evaluation is the focus of human resource management practice. Scientific evaluation and selection of management personnel is an effective way for enterprises to acquire core competitiveness. It is also the premise for enterprises to make full use of their talents and optimize the allocation of human resources. It is difficult to popularize and apply the evaluation index system because the weight of the evaluation index is too much depended on the expert experience and can not be combined with the actual situation of the enterprise. With the continuous advance and deepening of price research, various methods of talent evaluation have been applied to the process of comprehensive evaluation of talents. However, how to combine talent evaluation with the actual employment of enterprises, accord with the employing tendency of enterprises, combine enterprise talent evaluation with enterprise culture and development strategy, develop and construct suitable for enterprises themselves This research combs and summarizes the relevant theories of talent, management talent, talent evaluation at home and abroad, and the talent evaluation methods used in the current research. Combined with literature analysis, 17 influential enterprises are selected according to the results of frequency statistics. Based on the previous research results on talent, management talent theory and quality structure model, an evaluation index system of enterprise management talent is established, which contains four first-level indicators and 17 second-level indicators. Entropy method is used to determine the weight of the evaluation index, and the comprehensive score of the evaluation object is obtained by BP neural network modeling. Then BP neural network is constructed to establish the evaluation system of enterprise management personnel through network learning and testing. The learning sample set of the network is divided into two parts: one is the sample data obtained by the entropy method, which represents the experience of experts; the other is the sample data reflecting the actual employment and Employment Tendency of enterprises, and the salary level of enterprise management personnel is selected as the comprehensive evaluation score (salary water). Ping is a summary of the experience of enterprises in the process of selecting and evaluating talents for many years. It is a reflection of different employing tendencies of different enterprises and a practical result of matching talents and posts. The actual part of the learning sample collected by the enterprise can be used as a test sample set to test the accuracy of the neural network. Through the test of the neural network, the evaluation error of the test sample is less than 5%. This shows that the neural network model has good evaluation ability and can be used in the subsequent enterprise management. In this paper, BP neural network is used to evaluate enterprise management talents, which is not only a supplement to the existing evaluation methods of enterprise management talents, but also can avoid the evaluation deviation caused by subjective factors. This research establishes the enterprise management talent evaluation system, hoping to provide some reference for the selection and employment of enterprise management personnel.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP183;F272.92
【参考文献】
相关期刊论文 前10条
1 周利民;李晓蓉;李建红;朱琳;蒋玉鹤;;企业科技人才绩效管理评价模型构建研究[J];中国市场;2016年04期
2 杨文芳;;论企业管理的本质[J];现代经济信息;2016年02期
3 吴晓云;张欣妍;;企业能力、技术创新和价值网络合作创新与企业绩效[J];管理科学;2015年06期
4 张正堂;李瑞;;企业高绩效工作系统的内容结构与测量[J];管理世界;2015年05期
5 储祖旺;陈彪;刘芳雅;;我国高校学生事务管理者选聘机制优化研究[J];中国高教研究;2014年03期
6 田瑞雪;;国内图书馆数据挖掘技术应用研究述评[J];科技信息;2014年01期
7 王永丽;郑婉玉;;双重角色定位下的工会跨界职能履行及作用效果分析[J];管理世界;2012年10期
8 刘志成;吴能全;;中国企业家行为过程研究——来自近代中国企业家的考察[J];管理世界;2012年06期
9 陈俊杰;李中斌;;胜任能力模型研究述评[J];价值工程;2012年14期
10 蔡双立;;管理的本质与能力构建[J];现代财经(天津财经大学学报);2012年05期
相关博士学位论文 前2条
1 兰兰;东湖高新区高层次人才评价研究[D];华中科技大学;2014年
2 潘小炎;广西全科医生综合素质评价指标体系研究[D];中南大学;2013年
相关硕士学位论文 前8条
1 温必成;电网企业一般管理人员能力模型构建[D];华南理工大学;2015年
2 彭蕾;我国科技人才评价方法论研究[D];北京化工大学;2014年
3 王笑铮;基于胜任力模型的电子商务人才评价指标研究[D];太原科技大学;2014年
4 周鑫;基于模糊综合评判法的企业高级管理人才评价研究[D];北京邮电大学;2013年
5 张洪燕;基于熵值法和SEM的高层次外贸人才评价指标体系研究[D];江苏科技大学;2012年
6 聂燕飞;基于层次分析法和灰色关联分析的油田科技人才评价模型研究[D];上海交通大学;2011年
7 胡蓓蕾;论公司章程自治与界限[D];华东政法大学;2010年
8 王小梅;国际职业经理人素质测评研究[D];大连海事大学;2005年
，本文编号：2223538
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2223538.html
