融合LSTM与逻辑回归的中文专利关键词抽取  

Extracting Chinese Patent Keywords with LSTM and Logistic Regression

作　　者：韦婷婷 江涛[1] 郑舒玲 张建桃[1] Wei Tingting;Jiang Tao;Zheng Shuling;Zhang Jiantao(College of Mathematics and Informatics,South China Agricultural University,Guangzhou 510642,China;Patent Examination Cooperation Guangdong Center of the Patent Office,Guangzhou 510535,China)

机构地区：[1]华南农业大学数学与信息学院,广州510642 [2]国家知识产权局专利局专利审查协作广东中心,广州510535

出　　处：《数据分析与知识发现》2022年第2期308-317,共10页Data Analysis and Knowledge Discovery

基　　金：广东省普通高校青年创新人才类项目(项目编号:2019KQNCX012);广东省基础与应用基础研究区域联合青年基金项目(项目编号:2019A1515110396);教育部人文社会科学研究一般项目(项目编号:20YJC740067)的研究成果之一。

摘　　要：【目的】构建一种能够更加准确地抽取中文专利关键词的方法,解决传统方法难以发现低频、长尾关键词的问题。【方法】提出一种融合长短期记忆神经网络和逻辑回归模型的候选关键词抽取方法,进一步构建重组过滤规则,实现目标关键词抽取。【结果】实验结果表明,所提方法在所有关键词、低频关键词、长尾关键词、低频长尾关键词的抽取准确率比其他方法分别提高了5、24、11、26个百分点。【局限】所提方法通过设定阈值实现关键词的分类,对阈值附近点的词语划分仍不够精确。【结论】融合长短期记忆神经网络和逻辑回归模型获取候选关键词,并基于重组过滤规则提取目标关键词,能够有效发现专利文本中的低频、长字符组成的关键术语,为专利文本分析等任务提供一定的参考价值。[Objective]This paper constructs a new method to extract keywords from Chinese patents based on the LSTM and logistic regression,aiming to identify low-frequency and long-tail keywords effectively.[Methods]First,we combined the LSTM neural network and logistic regression model to extract the candidate keywords.Then,we reconstructed the filtering rules to retrieve the target keywords.[Results]The extraction accuracy of all keywords,low-frequency keywords,long-tail keywords,and low-frequency long-tail keywords were 5%,24%,11%and 26%higher than those of existing methods.[Limitations]The proposed model classifies keywords by setting thresholds,which are not precise to process words near the thresholds.[Conclusions]Our new model could effectively discover key terms with low frequency and long characters from texts,which benefits patent analysis and other services.

关 键 词：关键词抽取 LSTM神经网络 逻辑回归 重组过滤 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]