摘要
针对中小企业的发展特点,提出了从经营现状和发展潜力两个维度进行综合评价中小企业绩效的二维模型,并采用创业板中171家中小企业的数据进行了模型验证分析。考虑到评价企业数据量较大,论文依托于聚类分析方法,提出了一种典型样本企业选取策略,然后应用优势粗糙集理论对典型样本集进行了专家知识学习,形成中小企业绩效评价的决策规则,对所有企业进行绩效分类,构建出二维评价模型。其结果分析表明,我国中小企业在现状和发展潜力方面表现均优的企业较少,企业的发展潜力存在不足。此外,基于实际数据,论文讨论和演算了训练样本数量与粗糙集学习分类质量关系,发现粗糙集学习分类中存在过学习现象,即训练样本数的增多并不一定能提高分类质量。
Aiming at the development characteristics of small and medium-sized enterprises （SMEs）, a performance evaluation model in the dimension of management status and development potential for SMEs is put forward and the model is validated analytical with 171 SMEs data from gem. Taking into account the large number of businesses, a typical business sample selection policy relying on the cluster analysis method is presented, and then dominance rough set theory incorporated into the performance evaluation model to provide the decision rules generation for SMEs performance evaluation by expert knowledge learning. Thus, a two-dimensional performance evaluation model for SMEs is constructed. The evaluation results show that there are few SMEs with both excellent status quo and development potential. In addition, the relationship between the quality of rough set classification algorithm and the size of training samples is discussed. Results show that there is over fitting phenomena in learning classification of rough set, i. e. , increasing the number of training samples is not necessarily able to improve the learning quality of classification.
出处
《工业工程》
2015年第1期119-127,共9页
Industrial Engineering Journal
基金
国家自然科学基金面上资助项目(71471087)
南京航空航天大学基本业务基金产学研专项(2010-091-21L)
关键词
绩效评价
优势粗糙集
二维模型
分类质量
performance evaluation
dominance rough set theory
two-dimensional model
quality of rough set classification
