图书情报工作 ›› 2020, Vol. 64 ›› Issue (11): 87-95.DOI: 10.13266/j.issn.0252-3116.2020.11.010

• 知识组织 • 上一篇    下一篇

金融科技专利识别与分类方法构建及应用

徐璐, 卢小宾, 杨冠灿   

  1. 中国人民大学信息资源管理学院 北京 100872
  • 收稿日期:2019-12-23 修回日期:2020-03-14 出版日期:2020-06-05 发布日期:2020-06-05
  • 通讯作者: 卢小宾(ORCID:0000-0002-2866-4509),信息分析研究中心主任,教授,博士,博士生导师,通讯作者,E-mail:luxb@ruc.edu.cn
  • 作者简介:徐璐(ORCID:0000-0003-2414-3180),讲师,博士;杨冠灿(ORCID:0000-0002-1706-1884),讲师,博士。
  • 基金资助:
    本文系国家自然科学基金青年项目"银行业竞争政策的实施效果及影响机制研究"(项目编号:71903189)和中国博士后科学基金项目"多元所有制银行市场中强化竞争的风险和福利效应"(项目编号:2019M660052)研究成果之一。

Identify and Classify FinTech Patent

Xu Lu, Lu Xiaobin, Yang Guancan   

  1. School of Information Resource Management, Renmin University of China, Beijing 100872
  • Received:2019-12-23 Revised:2020-03-14 Online:2020-06-05 Published:2020-06-05

摘要: [目的/意义] 金融科技在信息和数据时代发展迅速,专利数目不断增长,同时金融科技的领域交叉和边界模糊等特性,也提升了其专利分析的难度。因此需要构建适合的识别与分类方法,从而准确高效地处理持续增长的大体量数据。[方法/过程] 首先依据金融科技的内涵和功能,梳理其包含的创新类别,厘清金融科技专利的范围和边界;随后使用机器学习算法,结合文本过滤、人工判读,构建金融科技专利识别与分类的方法流程。[结果/结论] 提出一种基于机器学习算法的专利识别与分类流程,能够较为准确高效地识别金融科技专利并进行分类,并通过分析所得的金融科技专利分类数据,总结当前金融科技的发展现状。

关键词: 金融科技, 信息经济学, 专利分析, 机器学习

Abstract: [Purpose/significance] FinTech has developed rapidly in the information and data era, and the number of patents has continued to increase. At the same time, its field crossover and blurred borders characteristics have also increased the difficulty of patent analysis. Therefore, it is necessary to construct a suitable identification and classification method, so as to accurately and efficiently process the continuously growing large volume data. [Method/process] This paper firstly sorted out the innovation categories based on the connotation and function of FinTech, and thus clarified the scope and boundaries of FinTech patents. Then, it constructed a FinTech patent recognition and classification model based on machine learning algorithms, combining text filtering and manual interpretation. [Result/conclusion] This paper proposes a patent recognition and classification process, based on machine learning algorithms, which is more accurately and efficiently. By analyzing the obtained FinTech patent classification data, the research also summarizes the current FinTech development status.

Key words: FinTech, information economics, patent analysis, machine learning

中图分类号: