专利网络中群体行为视角的科技创新趋势涌现研究

In the age of big data and big science, it has become a share focus issue of technology, industry and academia to sort out the context of knowledge development from heterogeneous information resources and making accurate judgments on the key areas and innovation trends of science and technology. We try to study the emerging mechanism and detecting methods of the trends of technological innovation in patent network from the perspective of collective behavior. Firstly, a large-scale patent network is constructed, which collected knowledge and information into a correlated system that extends infinitely, both temporally and spatially. Secondly, we explore the rule of knowledge creation activities at both temporal and spatial scales by making statistical analysis of the time interval and spatial displacement of consecutive knowledge creation activities, and explore the mapping between the technology base and technology trends through examining the learning rules of inventor groups. Then a collective dynamics model based on patent network has been build, which studies the process of group activity and emergence of innovation trends. At last, technology trends analysis system has been build, which include the functions of technology path highlighting and innovative trend detecting. This research has important implications for revealing the rule of creation activities of inventor group and illustrating the process of innovation trends emerging. We suppose our work could provide methods and tools for technological development and trends prediction, and provide decision-making reference for countries and enterprises to develop science and technology strategy.

大数据和大科学的时代背景下，从庞杂的信息资源中理清知识发展的脉络、对科技的重点领域和创新趋势做出准确的判断成为科技、产业和学术界所共同关注的问题。本课题尝试从群体行为视角研究专利网络中科技创新趋势的涌向机制和探测方法：通过构建大规模专利网络，将人类的发明创造构建为一个时间和结构上无限延展的关联系统；通过统计发明者群体创造性活动在时间上的密度和空间上的广度，研究发明者群体创造性活动的迁移轨迹；考察专利网络中的发明者群体的学习规则，探讨科技基础与科技趋势之间的映射关系；构建专利网络中的群体动力学模型，研究发明者群体一致行动的发生和创新趋势的涌向过程；设计科技路径凸显和创新趋势探测的算法，建立基于专利网络的创新趋势分析系统；该研究对于揭示发明者群体创新性行动的行为规则，阐明科技创新趋势的涌现过程具有重要的意义，可以为科技发展和趋势预测提供方法和工具，为国家和企业制定科技发展战略提供决策参考。

本课题通过构建大规模专利网络，将人类的发明创造构建为一个时间和结构上无限延展的关联系统；通过统计发明者群体创造性活动在时间上的密度和空间上的广度，研究发明者群体创造性活动的迁移轨迹；考察专利网络中的发明者群体的引证和学习规则，探讨科技基础与科技趋势之间的映射关系；构建专利网络中的群体动力学模型，研究发明者群体一致行动的发生和创新趋势的涌向过程；设计科技路径凸显和创新趋势探测的算法，建立了基于专利网络的创新趋势分析系统。截至目前为止，项目组合计完成学术论文12篇，出版专著1部；收集6,156,471条专利全文记录，建立了专利数据库；建立专利分析原型系统1个，具备WEB信息服务、专利检索、专利分析、专利网络构建、专利网络演化仿真和创新趋势探测几个方面的功能。