江苏四市政府绩效评估指标挖掘结果  

Mining Results of Government Performance Evaluation Indicators of Four Cities in Jiangsu

作　　者：尚虎平[1,2] 

机构地区：[1]兰州大学管理学院,兰州730000 [2]科技部中国科学技术信息研究所,北京100038

出　　处：《甘肃行政学院学报》2013年第1期10-24,126,共15页Journal of Gansu Administration Institute

基　　金：国家社科基金重大项目(编号:11&ZD057);国家教育部人文社会科学研究青年项目(编号:09YJC630126);国家自然科学基金(编号:70873100)的阶段性成果

摘　　要：在数据仓库基础上,本研究利用RBF神经网络"软聚类"挖掘出了在特定支持度下的10个一级指标,即经济发展、公民满意、人民生活改进、社会和谐、可持续发展、公共服务、内部过程、电子政务、人的发展、廉洁行政;在一级指标的基础上,进而利用BIRCH硬聚类思想,分层挖掘出了在特定支持度下的30个二级指标,即经济增长、产业协调、经济结构协调、抽象行政行为满意、具体行政行为满意、服务态度满意、生活改善、生活负担、参与机会、社会公平、社会救助、社会保障、生态建设、环境治理、低碳普及、科技文化发展、基础设施供给、保障性服务、人员素质、组织结构、管理流程、信息的丰富度、信息的及时度、反馈处理度、人口结构、教育程度、健康程度、实体性腐败、工作作风、行政效率等;在三级指标的基础上,仍然采用BIRCH分层硬聚类,产生了90个三级指标。在结果之外,我们还对三级指标进行了合用性丰富,为每个指标提供了未来实践评估中可以把我的指标标准,这使得这套指标可以被直接用之于政府的日常绩效评估中。In the process of this exploratory study,the paper firstly takes the advantage of the idea of Data Mining to define the key concept 'performance','Government Performance','Government Performance Evaluation'.Furthermore,it clarifies and explains 'Government Performance evaluation indicators' and 'Data Mining'.At last,it recalls achievements and Shortcomings of Government Performance evaluation indicators that have been designed at home and abroad.Based on the efforts,this paper starts substantial study on Government Performance evaluation indicator designing.In the study,the usability of Data Mining used in designing the government performance evaluation indicators is manifested first,and then,the exploratory study sites are chosen under certain needed conditions.On finishing these tasks,four municipal governments' performance indicator data warehouse is established,and a certain target domain(stair indicators) and threshold(Weight) are mined out of the warehouse using RBF neural network technology.Eventually,the secondary indicators,triple indicators and weight are clustered by BIRCH hierarchical clustering tool.In order to make the indicators more practical in the everyday realistic evaluation work,the triple indicators are enriched by some definitions,standards,and calculating formula.

关 键 词：数据挖掘 地方政府绩效评估 评估指标 挖掘结果 评估标准 

分 类 号：D035[政治法律—政治学]