新兴技术竞争情报挖掘方法研究

发布时间：2018-01-04 06:17

本文关键词：新兴技术竞争情报挖掘方法研究　出处：《北京理工大学》2016年博士论文　论文类型：学位论文

【摘要】：战略性新兴产业已成为当前全球经济一体化大趋势下,实现经济可持续发展,抢占国际竞争优势地位,提升国家核心竞争力的革命性力量。党的“十八大”明确提出了“实施创新驱动发展战略”的伟大战略思想,我国推动战略性新兴产业的重要举措全面展开。在这样的背景下,以“新兴技术竞争情报”为研究对象,构建“新兴技术竞争情报挖掘方法体系”,萃取新兴技术核心技术组件,发现新兴技术热点问题,定义新兴技术内在关系,整合新兴技术潜在价值,描绘新兴技术演化趋势,并定位新兴技术产业竞争与合作关系,为国家与产业研究与发展战略提供有效数据支撑,成为本文的核心研究目的。本论文围绕“新兴技术竞争情报采集与萃取”、“新兴技术竞争情报分析”以及“新兴技术竞争情报可视化”三部分研究展开,并选择中国“染敏太阳能电池”技术领域进行应用研究,其结论有效验证了本文方法体系的科学性与有效性。本文的主要创新成果如下:(1)构建了“新兴技术竞争情报采集与萃取方法”研究模型构建了包括“科技文献数据检索策略方法”与“主题词簇方法”在内的“新兴技术竞争情报采集与萃取方法”研究模型。该模型以定量分析为主,定性分析为辅,针对科技文献数据源以及新兴技术主题词的相关特征,综合运用词表、模糊语义以及关联规则等技术方法,在有效采集新兴技术科技文献数据的基础上,构建了针对新兴技术主题词的清洗、合并与聚类的研究方案,为有效萃取新兴技术核心技术组件提供了技术方法,奠定了“新兴技术竞争情报挖掘方法体系”的数据基础。(2)构建了“新兴技术竞争情报分析方法”研究模型以“数理统计”与“语义分析”为核心,构建了“新兴技术竞争情报分析方法”研究模型。一方面,构建“数据驱动”的K-Means文本聚类方法,实现了针对特定科技文献数据源的高精度文本聚类以及基于训练集的K值局部最优值自动选定。另一方面,基于“语义分析”视角,将“主语—行为—宾语”结构与TRIZ创新理论中的“矛盾矩阵”理论相结合,构建“基于‘问题与解决方案模型’的语义TRIZ方法”,为深度挖掘新兴技术内部潜在关系提供了思路与方法。两套方法共同构成了“新兴技术竞争情报分析方法”研究模型,是“新兴技术竞争情报挖掘方法体系”的重要组成部分。(3)构建了“新兴技术竞争情报可视化方法”研究模型构建了包括“技术路线图构造方法”与“技术路线图选择与评估准则”在内的“新兴技术竞争情报可视化方法”研究模型。一方面,构建“基于主题词与话题的技术路线图构造方法”刻画微观层面的技术细节,构建“基于‘问题与解决方案模型’的技术路线图构造方法”刻画技术组件间的内在关系,描绘问题与解决方案的演化趋势,构建“半自动化技术路线图构造方法”,应用模糊集理论,提升技术路线图构造方法的自动化特性。另一方面,“技术路线图选择与评估准则”归纳总结技术路线图的适用性,为实际应用提供参考。“新兴技术竞争情报可视化方法”研究模型是“新兴技术竞争情报挖掘方法体系”的核心内容。
[Abstract]:Strategic emerging industry has become the global trend of economic integration, to achieve sustainable economic development, to seize the international competitive advantage and enhance the core competitiveness of the national revolutionary force. The party's "big eighteen" clearly put forward the implementation of innovation driven development strategy of the great strategic thought in China, an important measure to promote the overall development of strategic emerging industries in this context, the "emerging technology competitive intelligence" as the research object, construct the "emerging technology competitive intelligence mining system, extraction of emerging technology core technology components, found the problem emerging hot technology, emerging technology definition of internal relations, the potential value of the integration of emerging technologies, describes the evolution of emerging technology trends, and competitive positioning with the emerging technology industry cooperation, to provide effective data support for national and industrial research and development strategy, as the core Heart research purposes. This paper focuses on "emerging technology competitive intelligence collection and extraction", "analysis of competitive intelligence of emerging technologies and emerging technology competitive intelligence visualization" the three part of the study, and select "China dye sensitized solar cell technology field of applied research, the conclusion verified the correctness and effectiveness of this method system. The main contributions of this paper are as follows: (1) constructs the research model of emerging technology competitive intelligence collection and extraction method" is established including "science and technology literature data retrieval method" and "subject cluster method", "new technology competitive intelligence collection and extraction method of the model based model. With quantitative analysis, qualitative analysis, according to the scientific literature data source and characteristics of emerging technology subject, the integrated use of vocabulary, semantic fuzziness and relevance The rules of technical method, based on the effective collection of emerging technology science and technology literature data, constructed for the cleaning of emerging technology topics, research plan merging and clustering, provides technical methods for effective extraction of emerging technology core technology components, has established "the data base technology competitive intelligence system (mining method". 2) constructed the "emerging technology competitive intelligence analysis method" to "Statistics" and "semantic analysis" as the core, constructs the research model for emerging technology competitive intelligence analysis. On the one hand, the K-Means text clustering method to build a "data driven", realized the high precision for specific text clustering technology the data source and the training set based on the K values of the local optimal value is automatically selected. On the other hand, based on the perspective of "semantic analysis", "subject behavior object" And in the TRIZ theory of "contradiction matrix" theory, "based on the construction of" problem and solution model of "semantic TRIZ" method, provides ideas and methods for deep mining the potential relationship between emerging technologies. Two methods constitute the "emerging technology competitive intelligence analysis model", "an important part of the emerging technology competitive intelligence system" mining method. (3) constructs the research model visualization method "new technology competitive intelligence constructs including" technology roadmap construction method "and" technology roadmap selection and evaluation criteria "," emerging technology competitive intelligence visualization model. On the one hand, construction of technical details of the technology roadmap construction method of "keywords and topics based on the characterization of the micro level, build" problems and solutions based on the model of ". Operation roadmap construction method "depicts the relationship between the evolution of technology components, describe problems and solutions of the construction of" semi automation technology roadmap construction method, the application of fuzzy set theory, automation characteristics to promote the technology roadmap construction method. On the other hand, the technology roadmap selection and evaluation criteria for induction summary of technology roadmap, to provide reference for the practical application. The research model visualization method of emerging technology competitive intelligence is the core content of "emerging technology competitive intelligence mining system".

【学位授予单位】：北京理工大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：G350

【参考文献】