互联网领域潜在高价值专利的智能识别方法研究  

Intelligent Identification Algorithm of Potentially High Value Patent in Internet Field

作　　者：白利敏 朱哲[1] 刘琳[1] 夏鹏[1] BAI Limin;ZHU Zhe;LIU Lin;XIA Peng(Patent Examination Cooperation Tianjin Center of the Patent Offi ce,CNIPA,Tianjin 300304)

机构地区：[1]国家知识产权局专利局专利审查协作天津中心,天津300304

出　　处：《中国发明与专利》2018年第11期28-32,共5页China Invention & Patent

摘　　要：目前,"高价值专利"成为了各类知识产权政府报告中的热门词汇,企业和国家都聚焦于高价值专利的识别和挖掘,如何从海量的数据库中识别得出较为客观准确的评价结果成为亟待解决的问题。通过对现有评价方法的研究,现有评价模型多为通用型方法和模型,未考虑到各个领域特点,评价方法并不能很好适应所有领域专利的价值评价;且现有方法主要关注授权保护的专利价值的评估,对于审查中潜在高价值专利的识别未有相应评价模型和方法。因此,本文以互联网领域为例,提出一种基于神经网络算法的潜在高价值专利智能识别方法,为企业和政府对高价值专利的挖掘与培育提供智能化的评价工具。At present,the concept of"high value patent"is becoming increasingly popular in the fi eld of intellectual property.More and more companies are focusing on the identifi cation and mining of high value patents.In the internet,it is an urgent problem to be solved that identify objective and accurate evaluation results from massive databases.The existing evaluation models are mostly general methods and models without considering the characteristics of each domain and focus on value assessment of authorized patents.However,it is not applicable to that identification of potentially high value in the examination.In this paper,we propose a potential high value patented intelligent identification algorithm based on neural network algorithms in the fi eld of internet,to provide intelligent evaluation tools for the mining and cultivation of high value patents for enterprises and governments.

关 键 词：高价值专利 互联网 领域特点 智能评价 神经网络 

分 类 号：G306[文化科学]