专利挖掘是技术预见的核心方法之一,可以获取到专利中蕴含的高价值技术情报,为政府制定战略、企业研发规划提供决策支持。针对专利文本挖掘领域,本文提出一种构建SAO链的规则,将专利文本使用SAO链进行特征表示,充分利用了技术的层次和内容两部分信息,使专利文本相似度计算更加准确,然后围绕该方法构建了一套科学合理的技术预见方法体系。最后,以工业机器人领域为例进行实证研究,通过分析专利网络得到了该领域的技术发展路径图,通过专利地图识别出了该领域的五个技术机遇,为企业和政府提供了可借鉴的情报信息。

Patent mining is one of the core methods of technology foresight, which can obtain high-value technical information contained in patents and provide decision-making support for the government to formulate strategies and enterprise R D plans. Aiming at the field of patent text mining, this paper puts forward a rule of constructing SAO chain, and using SAO chain for feature representation of patent text, and making full use of the technical level and content information, which makes the calculation of patent text similarity more accurate, then constructs a scientific and reasonable technology prediction method system around this method. Finally, an empirical study is carried out in the field of industrial robots. Through analyzing the patent network, the technology roadmap in this field is obtained. And five future technological opportunities in this field are identified through the patent map, which provides useful information for relevant enterprises and governments.

G350

广东省软科学重大项目“科技革命与技术预见智库建设”( 2016B070702001)；广东省自然科学基金自由申请项目“考虑不确定性及动态博弈的云监控服务定价与应用研究”(2018A030313079)