基于场景需求的专利技术组合识别——以人形机器人为例  

Patent Technology Combination Identification Based on Scenario Requirements:An Example in the Field of Humanoid Robots

作　　者：由丽萍[1] 刘云鹏 YOU LiPing;LIU YunPeng(School of Economics and Management,Shanxi University,Taiyuan 030031,P.R.China)

机构地区：[1]山西大学经济与管理学院,太原030031

出　　处：《数字图书馆论坛》2025年第2期12-22,共11页Digital Library Forum

基　　金：山西省科技战略研究专项“人工智能技术赋能山西省传统产业绿色转型发展对策研究”(编号:202304031401035)资助。

摘　　要：专利作为创新成果的标志,对于推动经济发展具有重要的作用,根据不同场景下的技术需求挖掘潜在的专利技术组合,可以更精准地促进专利技术的应用和商业化,提升专利转化效率。基于场景语义关系的方法,通过整合专利、文献数据,构建场景化技术功效知识图谱;从商业报告和用户评论数据中挖掘技术场景需求;从技术互补性、技术相似性和技术独特性3个维度出发,将技术功效与场景需求进行匹配,挖掘不同场景下具有市场潜力的专利技术组合。以人形机器人领域为例,通过实证研究识别出6个技术应用场景,及对应的场景需求和潜在的专利技术组合,证明了识别方法的有效性。Patents,as a hallmark of innovative achievements,play a crucial role in driving economic development.Mining potential patent technology combinations according to technical requirements in different scenarios can more precisely promote the application and commercialization of patent technologies while improving patent conversion efficiency.This study proposes a scenario-semantic relationship-based methodology that constructs a scenario-oriented technology-efficacy knowledge graph by integrating patent and literature data.It extracts technical scenario requirements from business reports and user reviews,and matches technological efficacy with scenario requirements through three dimensions:technological complementarity,technological similarity,and technological uniqueness,so as to identify market-potential patent technology combinations across scenarios.Taking humanoid robot as an example,empirical research identifies six technical application scenarios with corresponding scenario requirements and potential patent technology combinations,demonstrating the effectiveness of this identification approach.

关 键 词：场景语义 场景需求 专利技术组合 技术识别 

分 类 号：G255.53[文化科学—图书馆学]