论文总字数：29550字

摘 要

在信息化时代背景下，科研协同创新平台的搭建受到越来越多学者的关注。高校科研信息收集和统计的复杂程度使得科研协同创新平台的构建尤为重要。科研协同创新平台可以更新科研信息，通过可视化分析使得科研研究现状更加清晰，利用学者推荐可在一定程度上减少学者寻找合作的阻碍，促进科研合作创新。但是目前已经投入使用的协同创新平台却很少，而且已有平台中对学者推荐和可视化分析两方面并没有很多研究，本文将针对这两方面开展进一步研究。

本文以南京工业大学为例，首先收集了南京工业大学部分学者基本信息、科研、教学、社会工作以及其他信息数据作为科研协同创新平台搭建的基础测试数据，并以学科评估表为依据，以学院为单位对学者信息进行了统计分析。其次，为了减少学者寻找合作者的阻碍，在前人的研究基础上，使用python语言采用基于内容的推荐算法模拟了科研协同创新平台学者推荐实现过程，为科研协同创新平台实现学者推荐提供思路和案例。最后，为了说明平台在科研协同创新工作中发挥的作用，对科研协同创新平台进行了可视化分析，从个人科研管理、协同学者发现、学者推荐三个方面说明平台如何服务学者，从统计分析、主题词云、学者聚类三个方面说明平台如何服务于组织管理者。

本文学者推荐及可视化分析在科研协同创新平台均有体现，为科研协同创新平台构建提供理论依据以及实践参考。在理论贡献方面，利用基于内容的推荐算法总结出适合科研协同创新平台的学者推荐算法；在实践贡献方面，实现学者推荐案例，为今后高校协同创新平台构建提供参考。

关键词: 科研协同创新 学者推荐 推荐算法 可视化

Research Collaborative Innovation Platform Based on Scholar Portrait: Scholar Recommendation Implementation and Visual Analysis

Abstract

In the information age, more and more scholars pay attention to the construction of scientific research collaborative innovation platform. The complexity of scientific research information collection and statistics in universities makes the construction of scientific research collaborative innovation platform particularly important. Scientific research collaborative innovation platform can update scientific research information, make research status clearer through visual analysis, and use scholars’ recommendation can reduce the obstacles of scholars seeking cooperation to a certain extent, and promote scientific research collaborative innovation. However, few collaborative innovation platforms have been put into use, and there is not much research on scholars’ recommendation and visual analysis in the existing platforms. This paper will carry out further research on these two aspects.

Taking Nanjing University of Technology as an example, this paper firstly collects some basic information, scientific research, teaching, social work and other information data of some scholars of Nanjing University of Technology as the basic test data for the platform of scientific research collaborative innovation. Based on the subject evaluation table, the paper makes a statistical analysis of the scholar information in the College as a unit. Secondly, in order to reduce the obstacles for scholars to find collaborators, on the basis of previous studies, using Python language and content-based recommendation algorithm to simulate the implementation process of scholars’ recommendation in scientific research collaborative innovation platform, providing ideas and cases for scholars’ recommendation in scientific research collaborative innovation platform. Finally, in order to illustrate the role of the platform in scientific research collaborative innovation, a visual analysis of the platform is carried out. It explains how the platform serves scholars from three aspects: individual scientific research management, collaborative scholar discovery and scholar recommendation. It also explains how the platform serves organizational managers from three aspects: statistical analysis, college word cloud and scholar clustering.

In this paper, scholars’ recommendation and visual analysis are embodied in scientific research collaborative innovation platform, which provides theoretical basis and practical reference for the construction of scientific research collaborative innovation platform. In terms of theoretical contribution, the content-based recommendation algorithm is used to summarize the scholars’ recommendation algorithm suitable for scientific research collaborative innovation platform; in terms of practical contribution, scholars’ recommendation cases are realized to provide reference for the future construction of university collaborative innovation platform.

Keywords: scientific research collaborative innovation; Scholar recommendation; Recommendation algorithm; visualization

目 录

摘 要 I

Abstract II

第一章 绪论 1

1.1 研究背景 1

1.2 文献综述 3

1.2.1 国内研究现状 4

1.2.2 国外研究现状 7

1.3 研究意义 9

1.4 论文结构 9

第二章 平台测试数据收集及分析 11

2.1 数据概述 11

2.2 数据收集 11

2.2.1 数据来源 11

2.2.2 数据收集字段 12

2.2.3 收集数据 13

2.3 数据分析 15

2.4 本章小结 21

第三章 科研协同创新平台的学者推荐算法及系统实现 22

3.1 学者推荐算法概述 22

3.2 学者推荐算法步骤 22

3.3 学者推荐算法系统实现 24

3.3.1 相似度计算案例 24

3.3.2 学者推荐案例 26

3.4 本章小结 30

第四章 科研协同创新平台可视化分析 31

4.1 平台概况分析 31

4.2 面向学者服务的科研协同创新平台分析 31

4.2.1 个人科研管理 32

4.2.2 协同学者发现 32

4.2.3 学者推荐 34

4.2.4 案例分析 35

4.3 面向组织管理的科研协同创新平台分析 35

4.3.1 统计分析 35

4.3.2 主题词云 37

4.3.3 学者聚类 38

4.3.4 案例分析 39

4.4 本章小结 39

第五章 总结与展望 40

5.1 总结 40

5.2 展望 40

参考文献 41

附录1 第四轮学科评估指标体系（二） 45

附录2 学者推荐案例代码 46

请支付后下载全文，论文总字数：29550字