图书情报工作 ›› 2021, Vol. 65 ›› Issue (5): 118-125.DOI: 10.13266/j.issn.0252-3116.2021.05.012

• 情报研究 • 上一篇    下一篇

高校可转移专利识别研究——基于贝叶斯理论和组合赋权法

韩盟1, 吴红1, 李昌2, 崔哲1, 李剑飞1   

  1. 1. 山东理工大学信息管理研究院 淄博 255049;
    2. 大连理工大学科学学与科技管理研究所 大连 116024
  • 收稿日期:2020-08-04 修回日期:2020-10-19 出版日期:2021-03-05 发布日期:2021-04-14
  • 通讯作者: 吴红(ORCID:0000-0002-1708-7638),研究馆员,硕士生导师,通讯作者,E-mail:wuhong0256@163.com
  • 作者简介:韩盟(ORCID:0000-0002-1532-9146),硕士研究生;李昌(ORCID:0000-0002-2454-792X),博士研究生;崔哲(ORCID:0000-0002-6803-2151),硕士研究生;李剑飞(ORCID:0000-0001-7884-2126),硕士研究生。
  • 基金资助:
    本文系国家社会科学基金项目"高校图书馆深度嵌入专利运营研究"(项目编号:16BTQ029)研究成果之一。

Identification of University Transferable Patent: Based on Bayesian Theory and Combination Weighting Method

Han Meng1, Wu Hong1, Li Chang2, Cui Zhe1, Li Jianfei1   

  1. 1 Institute of Information Management, Shandong University of Technology, Zibo 255049;
    2 Institute of Science and Technology Management, Dalian University of Technology, Dalian 116024
  • Received:2020-08-04 Revised:2020-10-19 Online:2021-03-05 Published:2021-04-14

摘要: [目的/意义] 研究高校可转移专利的识别,对于提高专利推送质量,促进高校科研与社会经济的对接具有积极意义。[方法/过程] 首先在文献调研的基础上检验并确定可量化专利识别指标,并结合贝叶斯理论对高校可转移专利进行初步筛选;然后使用复相关系数-变异系数组合赋权法计算各识别指标权重,并计算剩余专利的加权综合转移概率;最后依照综合概率值大小识别可转移专利,并用转移数量较高的医用配置品领域高校专利对本方法进行实证。[结果/结论] 该方法利用贝叶斯理论和组合赋权方法,通过初步筛选和二次识别,实现了对高校可转移专利综合概率的有效计算,即保证了结果的准确性,又兼顾到高校专利管理资源的有限性,为提高专利推送质量奠定了良好的基础。

关键词: 专利转移, 识别方法, 贝叶斯理论, 组合赋权

Abstract: [Purpose/significance] Research on the identification method of University transferable patents is positive significance to improve the quality of patent push and promote the docking of university scientific research and social economy. [Method/process] Firstly, based on the literature research, we tested and determined the quantifiable identification index of patents, and combined with Bayesian theory to screen university transferable patents preliminarily. Then, we used the multiple correlation coefficient and coefficient of variation combination weighting method to calculate the weight of each identification index, and calculated the weighted comprehensive transfer probability of the remaining patents. Finally, we identified transferable patents according to the comprehensive probability value. In addition, we used university patents in the field of medical configuration products to test the method. [Result/conclusion] We use Bayesian theory and combination weighting method to calculate the comprehensive probability of university transferable patents through preliminary screening and secondary identification, this identification process not only ensures the accuracy of the results, but also takes into account the limitation of university patent management resources, which builds a good foundation for improving the quality of patent push.

Key words: patent transfer, recognition methods, Bayesian theory, combination weighting

中图分类号: