我国高校专利分类变化的评价模型及其回归分析  

Evaluation Model and Regression Analysis of Patent Classification Changes in Chinese Universities

作　　者：谢雨飞[1] 朱周力 翟晓亮[1] 马琳[1] 李静玉[2] 朱平[2] 

机构地区：[1]江南大学物联网工程学院,江苏无锡214122 [2]江南大学理学院,江苏无锡214122

出　　处：《技术与创新管理》2014年第3期205-213,共9页Technology and Innovation Management

基　　金：国家自然科学基金资助项目(61300150;71271029);中央高校基本科研业务费专项资金资助项目(JUSRP51317B);国家大学生创新训练项目(201310295050)

摘　　要：通过文献检索和网络调查等定性分析方法梳理出我国高校专利申请量的所有可能的影响因素,并以2001-2011年为时间轴,采集我国高校各类专利申请数量及其影响因素的相关数据。绘制基于Multi-Agent技术的专利地图,描绘出我国高校专利分类申请量的现状以及我国各地区高校专利申请量的情况。利用多元线性回归分析中的逐步回归方法,构建专利申请量的预测模型并与实际数据进行拟合比较,充分验证了该模型的有效性并且根据该模型及其分析给出了国家专利发展战略的有效建议和方案。Some qualitative analysis methods such as literature research and network investigation are applied to find out allthe possible factors influencing patent applications in Chinese Universities. The paper collected all related data of patent ap-plications of various types in Chinese Universities and their influencing factors in the period of 2001--2011, drew a multi-a-gent technology patent map, depicting the current situation of the different kinds of the patents applications in Chinese uni-versities and universities patent applications in different regions. Using the stepwise regression method of multiple linear re-gression analysis, the predictive models of patent applications were built and the actual data were fitted with the comparisonfully validate the model, and finally, effective proposals and programs were given to the national patent development strategy.

关 键 词：专利分类Multi-Agent系统 专利地图 回归分析 预测模型 

分 类 号：G30[文化科学]