基于专利数据的技术融合研究综述

目的/意义] 基于专利数据开展技术融合研究是技术融合研究的主要方式和热点方向，面向为技术融合后续研究工作开展提供参考和启发的目标，对国内外基于专利数据的技术融合研究现状进行较为全面的综述。方法/过程] 将已有研究按研究内容划分为技术融合测度及预测方法研究、特定领域/多领域间/全领域的技术融合态势测度及趋势预测、技术融合的测度指标研究、影响技术融合的特征因素研究、基于技术融合视角进行技术机会发现以及技术融合与创新的关系研究六类，并对各类别取得的研究成果进行评述。结果/结论] 基于专利数据开展技术融合研究已经取得一定的成果，但是仍旧存在测度依据不合理、预测方法缺乏验证、全领域研究关注度低的问题，这些问题可通过引入语义关系优化技术融合网络、采用图神经网络技术改进技术融合预测方法、完善全领域技术融合测度和预测方法来解决。

Research Review on Technology Convergence Based on Patent Data

Purpose/significance]Technology convergence research based on patent data is the main way and hot direction of technology convergence research.In order to support following research work of technology convergence,this paper reviews the current research of technology convergence based on patent data.Method/process]According to the research content,the existing research can be divided into six categories:the research on the measurement and prediction method of technology convergence;the measurement and trend prediction of specific field,multi fields and whole field technology convergence;the research on the measurement index of technology convergence;the research on the characteristic factors influencing the technology convergence;the technology opportunity discovery based on the perspective of technology convergence and the research on the relationship between technology convergence and innovation.The research results were also commented.Result/conclusion]Research on technology convergence based on patent data has achieved certain results,but there are still some problems,such as unreasonable measurement basis,lack of validation of prediction methods and low attention of the whole field research.These problems can be solved by introducing semantic relationship to optimize technology convergence network,using GNN technology to improve technology convergence prediction method and improving technology convergence measurement and prediction in the whole field.