摘要
基于专利文献度量技术间关联,不仅应考虑技术间共现关系,还不可忽视技术间固有联系,单择其一作为度量标准会造成分析结果的片面性。文章以德温特手工代码(DMC)表征技术,通过计算DMC间的概念语义相似度来度量技术与技术间的固有联系,以DMC在专利文献中的共现关系来度量技术间共现关系,将技术间固有联系及其共现关系全面引入到技术分析中,提出基于DMC的技术分析方法。实验结果的合理性得到提升,将结果以技术图谱的形式可视化,图谱结构清晰,分类合理,能较好观察领域研究热点。
In terms of the correlation among the measurement technologies of patent documents, scholars should not only consider the technologies co-occurrence, but also take into account of the inherent link among technologies. Otherwise, the result will not be comprehensive. This paper uses Derwent Manual Code ( DMC ) to represent technology. The concepts semantic similarity a- mong DMC is used to measure the inherent finks among technologies and the co-occurrence of DMC in patent documents is used to measure the co-occurrence of technologies. Then by introducing both of the inherent links and co-occurrence into technical analysis, the paper proposes the technical analysis method based on DMC. The results are more reasonable with this method. The paper visualizes the resuhs, and it can be seen that the structure and the classification of the spectrum both improve a lot, which makes it easi- er to observe the research hot spots.
出处
《情报理论与实践》
CSSCI
北大核心
2017年第10期96-99,105,共5页
Information Studies:Theory & Application
关键词
专利文献
技术
语义相似度
共现分析
patent document
technology
semantic similarity
co-occurrence analysis