基于区间数证据分组合成的高校创新能力评价
发布时间：2019-02-16 07:00
【摘要】：高校创新能力关乎国家和高校自身的发展,对于存在高度不确定性的高校创新能力评价问题,本文提出了一种基于区间数证据分组合成的评价模型。首先,构建高校创新能力指标体系并确定指标权重。其次,采用核密度估计得到区间数分布形式的综合证据可信度矩阵,采用Wasserstein距离并综合考虑证据源权重和证据体间差异得出新的证据综合距离进行分组。再次,进行组内、组间的证据合成得到最终的评价结果。最后,通过对安徽某高校的实证分析,得到合理的预期结果,验证了该方法的有效性。
[Abstract]:The innovation ability of colleges and universities is related to the development of the country and universities themselves. This paper presents an evaluation model based on interval number evidence grouping composition for the evaluation of innovation ability of colleges and universities with high uncertainty. First of all, construct the innovation ability index system and determine the index weight. Secondly, the kernel density estimation is used to obtain the comprehensive evidence confidence matrix in the form of interval number distribution, and the Wasserstein distance is used to combine the weight of the evidence source and the difference between the evidence bodies to group the new evidence synthesis distance. Thirdly, the final evaluation results were obtained by the intra-group and inter-group evidence synthesis. Finally, the validity of this method is verified by an empirical analysis of a university in Anhui province, and a reasonable expected result is obtained.
【作者单位】： 合肥工业大学科学技术研究院;合肥工业大学管理学院;
【基金】：国家自然科学基金面上项目,基于文本情感和异质网络分析的社会化推荐研究(71471054,起止时间:2015.01-2018.12);国家自然科学基金重大研究计划培育项目,科研社交网络中融合多源异构大数据的科研兴趣图谱和智能推荐研究(91646111,起止时间:2017.01-2019.12)
【分类号】：G644
本文编号：2424171
[Abstract]:The innovation ability of colleges and universities is related to the development of the country and universities themselves. This paper presents an evaluation model based on interval number evidence grouping composition for the evaluation of innovation ability of colleges and universities with high uncertainty. First of all, construct the innovation ability index system and determine the index weight. Secondly, the kernel density estimation is used to obtain the comprehensive evidence confidence matrix in the form of interval number distribution, and the Wasserstein distance is used to combine the weight of the evidence source and the difference between the evidence bodies to group the new evidence synthesis distance. Thirdly, the final evaluation results were obtained by the intra-group and inter-group evidence synthesis. Finally, the validity of this method is verified by an empirical analysis of a university in Anhui province, and a reasonable expected result is obtained.
【作者单位】： 合肥工业大学科学技术研究院;合肥工业大学管理学院;
【基金】：国家自然科学基金面上项目,基于文本情感和异质网络分析的社会化推荐研究(71471054,起止时间:2015.01-2018.12);国家自然科学基金重大研究计划培育项目,科研社交网络中融合多源异构大数据的科研兴趣图谱和智能推荐研究(91646111,起止时间:2017.01-2019.12)
【分类号】：G644
【相似文献】
相关期刊论文 前2条
1 张美恋;;区间数多属性决策在课堂教学质量评价中应用[J];宁德师范学院学报(自然科学版);2014年02期
2 ;[J];;年期
，本文编号：2424171
本文链接：https://www.wllwen.com/jiaoyulunwen/gaodengjiaoyulunwen/2424171.html
