以市场需求为导向的企业技术研发方向识别研究  

The Identification of R&D Direction of Enterprises'Technology Oriented by Market Demand

作　　者：张硕 乔亚丽 李荣荣 Zhang Shuo;Qiao Yali;Li Rongrong(School of Economics and Management,Beijing University of Technology,Beijing 100124,China;School of Public Policy and Management,Tsinghua University,Beijing 100084,China;School of Economics and Man-agement,China University of Petroleum(East China),Qingdao 266580,China)

机构地区：[1]北京工业大学经济与管理学院,北京100124 [2]清华大学公共管理学院,北京100084 [3]中国石油大学(华东)经济管理学院,山东青岛266580

出　　处：《创新科技》2024年第7期63-74,F0002,共13页Innovation science and technology

基　　金：国家自然科学基金青年项目“颠覆性低碳技术识别及创新路径动态选择研究”(72304023);国家自然科学基金青年项目“生物医学领域新兴主题的识别和预测方法研究”(72104246);中国博士后科学基金面上项目“基于多数据源融合的颠覆性低碳技术演化路径识别与预测研究”(2023M730153);中国博士后科学基金面上项目“全球科创中心高端科技人才流动的时空演化规律及形成机制研究”(2023M741901);北京市朝阳区博士后工作经费资助项目“绿色能源领域‘卡脖子’关键核心技术甄选机制及突破路径研究”(2023ZZ-019)。

摘　　要：识别领域内重点技术研发方向,是企业技术创新活动中不可或缺的关键环节,不仅关乎资源优化配置和创新效率提升,而且是推动企业持续发展、实现技术突破的重要战略举措。面向市场(或用户)需求,选择从竞争者导向角度出发,构建企业技术研发方向识别模型。首先,利用文本挖掘技术对领域内相关专利文本进行SAO语义结构提取和专利权人提取,并以与技术相关的S或O概念层级实体为技术表征,构建企业—技术关联网络及企业合作网络;其次,将企业—技术关联矩阵进行Z-score标准化得到企业—技术评分矩阵,并将社团划分算法应用于企业合作网络,综合社团属性和企业相似性分值选取Top-k项企业所涉及的技术,形成候选技术列表;最后,构建技术增长性和技术适用性指标,对识别出的候选技术进行定量评估,并结合企业自身技术布局为其推荐适合的发展方向。实证研究部分,以默沙东公司(MERI-C)为例,进行技术研发方向识别研究,为企业精准高效开展技术创新活动提供科学的决策依据。How to identify R&D directions of key technologies within a given field is a pivotal element of enterprises'technological innovation activities.It not only concerns the optimal allo-cation of resources and the enhancement of innovation efficiency but also serves as a vital strate-gic measure for driving sustainable corporate growth and achieving technological breakthroughs.Given the nonlinear interactions among corporate strategies,market demands,and market envi-ronment,even satisfied market demands can present development opportunities for emerging en-terprises.This study constructs a model for identifying enterprises'technology R&D directions from a competitor-oriented perspective.First,text mining techniques are employed to extract Subject-Action-Object structures and patent assignees from domain-related patent texts.The extracted Subject/Object conceptual-level entities related to technology are used as technological repre-sentations to construct a firm-technology association network and a firm collaboration network.Second,the firm-technology association matrix is standardized using Z-score normalization to obtain a firm-technology scoring matrix.A community detection algorithm is then applied to the firm collaboration network,and the Top-k technologies involving firms are selected based on comprehensive community attributes and firm similarity scores,forming a candidate technology list.Finally,indicators for technological growth potential and applicability are constructed to quantitatively evaluate the identified candidate technologies,and recommended R&D directions are proposed for firms based on their existing technological layouts.Taking Merck as an example,this study identifies enterprises'technology R&D directions.The results indicate that tyrosine kinase inhibitors and BRAF inhibitors can be two major techno-logical areas of focus for Merck's future independent or collaborative R&D.From a technological standpoint,both tyrosine kinase inhibitors and BRAF inhibitors are means to fulfill existing de-mands

关 键 词：技术机会 SAO语义挖掘 协同过滤算法 技术研发方向 

分 类 号：G250[文化科学—图书馆学] G353