科研社交网络平台中的合作者推荐
发布时间:2021-11-26 07:21
寻求合作者是科研工作者的重要学术活动之一,因为合适的合作者会有助于提高学者的研究质量,加快其研究进程。随着信息技术的快速发展,科研社交平台已经广泛出现,并且吸引了大量的研究人员通过虚拟社区来开展科研合作。因此,在这些科研在线平台中开发出高效的合作者推荐系统将有效地促进学术合作与知识共享。信息过载和信息不对称是在合作者推荐的研究领域需要解决的两个关键问题。总的来说,本研究首先要定义出潜在的学术合作者推荐的情境,同时需要给出对应的解决方案,为用户提供有效的建议和决策支持。现有的合作者推荐研究主要关注研究者之间的相似度,如基于专业知识背景的相似性和社交网络的邻近度等。尽管在这一领域已经有很多的研究,但是对于科研合作者推荐的总体框架和有效的推荐算法仍然是缺乏的。在本研究中,我们提出了一个总体框架来解决科研合作者的推荐问题。该框架定义出了两个主要的合作者推荐情境,即基于相似性的合作者推荐,和在一个特定的背景限制下的合作者推荐。针对这两个推荐情境,本文提出了两个对应的高效的解决方案。对于基于相似性的合作者推荐问题,我们提出了一个混合方法,分别从专业知识的相关性、社交网络的邻近度和机构层面的合作度三...
【文章来源】:中国科学技术大学安徽省 211工程院校 985工程院校
【文章页数】:109 页
【学位级别】:博士
【文章目录】:
摘要
ABSTRACT
1 INTRODUCTION
1.1 Background and Motivation
1.2 Research Objectives
1.3 Research Approach
1.4 Research Organization
2 LITERATURE REVIEW
2.1 Academic Collaboration
2.2 Collaborator Recommendation Approaches
2.2.1 Content-based collaborator recommendations approaches
2.2.2 Homogeneous network-based collaborator recommendations approaches
2.2.3 Heterogeneous network-based collaborator recommendations approaches
2.3 Semantic Analysis Techniques
2.3.1 Probabilistic latent semantic indexing
2.3.2 Latent Dirichlet Allocation
2.3.3 Author-topic model
2.4 Social Network Analysis Techniques
2.4.1 Neighborhood-based network proximity measures
2.4.2 Path-based network proximity measures
2.4.3 Topology-based network proximity measures
2.5 Context-constrained Recommendation Approaches
2.6 Summary and Research Gaps
3 PROPOSED APPROACH
3.1 A General Collaborator Recommendation Framework
3.2 Similarity-based Collaborator Recommendation
3.2.1 The similarity-based collaborator recommendation framework
3.2.2 Expertise relevance
3.2.3 Social network proximity
3.2.4 Institutional connectivity
3.2.5 Fusing strategy
3.3 Collaborator Recommendation within a Specific Context
3.3.1 Recommendation framework
3.3.2 Expert quality profiling
3.3.3 Quality-weighted LDA model
3.3.4 Expertise coverage-oriented matching
4 EXPERIMENTAL EVALUATIONS
4.1 Evaluation of Similarity-based Collaborator Recommendation
4.1.1 Dataset preparation
4.1.2 Experimental design
4.1.3 Evaluation metrics
4.1.4 Results and discussions
4.2 Evaluation of Collaborator Recommendation within a Specific Context
4.2.1 Dataset construction
4.2.2 Experimental design
4.2.3 Evaluation metrics
4.2.4 Results and discussions
4.3 System Implementation of the Two Proposed Approaches
4.3.1 System implementation for similarity-based collaborator recommendation
4.3.2 System implementation for collaborator recommendation within a specific context
5 CONCLUSIONS AND FUTURE WORK
5.1 Summary of the Research
5.2 Contributions of the Research
5.3 Limitations of the Research
5.4 Future Research
REFERENCES
致谢
在读期间发表的学术论文与取得的其他研究成果
本文编号:3519661
【文章来源】:中国科学技术大学安徽省 211工程院校 985工程院校
【文章页数】:109 页
【学位级别】:博士
【文章目录】:
摘要
ABSTRACT
1 INTRODUCTION
1.1 Background and Motivation
1.2 Research Objectives
1.3 Research Approach
1.4 Research Organization
2 LITERATURE REVIEW
2.1 Academic Collaboration
2.2 Collaborator Recommendation Approaches
2.2.1 Content-based collaborator recommendations approaches
2.2.2 Homogeneous network-based collaborator recommendations approaches
2.2.3 Heterogeneous network-based collaborator recommendations approaches
2.3 Semantic Analysis Techniques
2.3.1 Probabilistic latent semantic indexing
2.3.2 Latent Dirichlet Allocation
2.3.3 Author-topic model
2.4 Social Network Analysis Techniques
2.4.1 Neighborhood-based network proximity measures
2.4.2 Path-based network proximity measures
2.4.3 Topology-based network proximity measures
2.5 Context-constrained Recommendation Approaches
2.6 Summary and Research Gaps
3 PROPOSED APPROACH
3.1 A General Collaborator Recommendation Framework
3.2 Similarity-based Collaborator Recommendation
3.2.1 The similarity-based collaborator recommendation framework
3.2.2 Expertise relevance
3.2.3 Social network proximity
3.2.4 Institutional connectivity
3.2.5 Fusing strategy
3.3 Collaborator Recommendation within a Specific Context
3.3.1 Recommendation framework
3.3.2 Expert quality profiling
3.3.3 Quality-weighted LDA model
3.3.4 Expertise coverage-oriented matching
4 EXPERIMENTAL EVALUATIONS
4.1 Evaluation of Similarity-based Collaborator Recommendation
4.1.1 Dataset preparation
4.1.2 Experimental design
4.1.3 Evaluation metrics
4.1.4 Results and discussions
4.2 Evaluation of Collaborator Recommendation within a Specific Context
4.2.1 Dataset construction
4.2.2 Experimental design
4.2.3 Evaluation metrics
4.2.4 Results and discussions
4.3 System Implementation of the Two Proposed Approaches
4.3.1 System implementation for similarity-based collaborator recommendation
4.3.2 System implementation for collaborator recommendation within a specific context
5 CONCLUSIONS AND FUTURE WORK
5.1 Summary of the Research
5.2 Contributions of the Research
5.3 Limitations of the Research
5.4 Future Research
REFERENCES
致谢
在读期间发表的学术论文与取得的其他研究成果
本文编号:3519661
本文链接:https://www.wllwen.com/shoufeilunwen/jjglss/3519661.html