基于SAO语义挖掘的技术研发合作伙伴识别与选择方法研究

随着技术创新知识复杂性的不断增加和产品生命周期的不断缩短，企业需要有目的地运用内外部知识流以拓展自身的技术知识边界，从而有效提升自身的研发能力，因此技术研发合作成为众多企业快速提升技术创新能力的优先选择之一。但企业如何有效地从海量数据中提取价值，快速识别和选择技术研发合作伙伴成为学术界和实业界长期关注的焦点。.本项目针对开放式创新背景下企业加强技术研发合作的现实需求，就如何有效利用海量科技文本数据识别和选择合适的技术研发合作伙伴问题开展研究，实现SAO语义结构的识别和优化，探索基于SAO语义结构的技术相似性和互补性测度，构建基于技术相似度和技术互补度的合作伙伴识别模型，在此基础上实现基于技术合作环境分析的合作伙伴最优选择，为不同机构开展技术研发合作创新，提升合作绩效提供决策指导，进一步推进国家创新体系建设。

With the increasing complexity of knowledge of technology innovation and shortening of the product's lifecycle, the enterprise needs to take advantage of internal and external knowledge flow to expand their technical knowledge boundary effectively and then to extend their R&D capabilities. However, how to effectively extract the value of science and technology from the volumes of text data,then to identify and select technological R&D partners has been drawing much attention from academia and industry. For the realistic demand to strengthen technology R&D cooperation in open innovation, this project conducts the research on how to use massive text data to identify and select appropriate technological R&D partners. This project achieves the identification and optimization of SAO semantic structure, proposes the SAO semantic structure-based measure methods of technology similarity and technology complementarity, and builds the R&D partners identification model based on technology similarity and technology complementarity, further realizes to select the optimal R&D partners by combining with technology cooperation environment analysis. Such results provide guidelines for different institutions' decision-making to carry out technology R&D cooperation innovation and promote cooperation performance.