基于BP神经网络的高校社科科研绩效评价体系研究

发布时间：2018-01-18 15:38

本文关键词：基于BP神经网络的高校社科科研绩效评价体系研究　出处：《南京理工大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着知识经济的到来,科教兴国已经成为全国上下的共识,各级政府和高等学校都加大了对科技的投入。因此,如何评价科研基金投资效益、如何评价高校科研人员工作完成情况,也就成为高校科研管理部门管理人员必须认真考虑的问题。目前,对高校的社科类科研绩效评价较为少见,另外,BP神经网络评价方法在我国才刚刚起步,具有一定的先进性。本文以实现理论研究与实证研究相结合、定性分析与定量分析相结合的研究方法,对高校社科类科研绩效评价进行深入的探讨和研究。首先介绍了论文研究背景及目的；其次在对高校科研进行总结和综述基础上,根据全面简明、客观可比、可操作性等设置原则构建高校社科类科研评价指标体系,引入BP神经网络方法；最后通过实证研究,包括研究所需数据的收集与处理和样本学习与仿真过程；在结尾部分对本文的研究结果进行总结,并对今后的高校社科类科研绩效评价的研究进行展望。本文研究最终采用Levenberg-Marquardt算法对网络进行了改进,建立评价模型并结合MATLAB软件实现评价过程。期望本文的研究方法及结论能够为评价高校社科类科研绩效提供依据。
[Abstract]:With the coming of knowledge economy, it has become the consensus of the whole country to invigorate the country through science and education. Governments at all levels and colleges and universities have increased their investment in science and technology. Therefore, how to evaluate the investment benefits of scientific research funds. How to evaluate the completion of scientific research personnel in colleges and universities has become a problem that managers of scientific research management departments must seriously consider. At present, it is rare to evaluate the performance of social science research in colleges and universities, in addition. BP neural network evaluation method has just started in our country, and has some advanced nature. In this paper, the realization of theoretical research and empirical research, qualitative analysis and quantitative analysis of the combination of research methods. This paper makes a deep discussion and research on the performance evaluation of social science research in colleges and universities. Firstly, it introduces the research background and purpose of the thesis. Secondly, on the basis of summing up and summarizing the scientific research in colleges and universities, the evaluation index system of social science research in colleges and universities is constructed according to the setting principles of comprehensive, concise, objective and comparable, and the BP neural network method is introduced. Finally, through the empirical research, including the research data collection and processing, sample learning and simulation process; At the end of this paper, the research results are summarized, and the future research on the performance evaluation of social science research in colleges and universities is prospected. In this paper, the Levenberg-Marquardt algorithm is used to improve the network. The evaluation model is established and the evaluation process is realized with MATLAB software. It is expected that the research methods and conclusions of this paper can provide the basis for evaluating the achievements of social science research in colleges and universities.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：G644

【参考文献】