Abstract: Identifying potential technology opportunities from the perspective of market demand and technological capability is crucial for enterprises to gain a first-mover advantage in the market.In view of the high uncertainty of market demand and the fact that enterprises often ignore their own technical capability in the research of market-driven technological opportunity identification,a new approach to identify technological opportunities from the dual perspective of market demand-enterprise technical capability was proposed.On the basis of collecting and preprocessing the data of technology and demand in the concrete domain,the Latent Dirichlet Allocation(LDA)topic model was used to extract the technology and market demand topics respectively,and the Word to vector(Word2vec)model was used to vectorize them.The collaborative technology topic based on association rules mining analyzed the technical ability of the target enterprise.The community discovery algorithm was employed to coordinate the market demand topic to achieve the mining of market demand.The similarity between enterprise technical capability and market demand was calculated by cosine similarity,and then the technical knowledge meeting the market demand was extracted.Based on this,the technical knowledge network was constructed,and the link prediction algorithm identified the potential technical opportunities in the network.Finally,the new energy vehicle battery technology was taken as an example to verify the feasibility of the proposed method.The application results showed that the technology opportunity identification path under the dual perspective of market demand-enterprise technical capability could provide a theoretical reference for enterprises in the field of concrete technology to carry out technological innovation activities efficiently.

Key words: technology opportunity identification, market demand, enterprise technical capability, association rules mining, link prediction

摘要： 针对市场驱动的技术机会识别研究中因市场需求的高度不确定性,以及企业常常忽视自身技术能力等问题,提出一种市场需求—企业技术能力双重视域下的技术机会识别路径。在搜集具象领域技术及需求数据并进行预处理的基础上,运用潜在狄利克雷分布主题模型分别提取了技术和市场需求主题,并利用词向量模型将其向量化;基于关联规则挖掘的协同技术主题剖析了目标企业的技术能力,并利用社区发现算法协同市场需求主题,以挖掘市场需求;通过余弦相似度计算企业技术能力和市场需求之间的相似性,进而提取市场需求对应的技术知识,依此构建技术知识网络,利用链路预测算法识别网络中的潜在技术机会。最后以新能源汽车电池技术为例验证了所提方法的可行性。

关键词: 技术机会识别, 市场需求, 企业技术能力, 关联规则挖掘, 链路预测