基于ATT-IndRNN-CNN的维吾尔语名词指代消解  

Anaphora Resolution of Uyghur Nouns Based on ATT-IndRNN-CNN

作　　者：祁青山 田生伟[1] 禹龙 艾山·吾买尔[2] QI Qingshan;TIAN Shengwei;YU Long;AISHAN Wumaier(School of Software,Xinjiang University,Urumqi,Xinjiang 830091,China;College of Information Science and Engineering,Xinjiang University,Urumqi,Xinjiang 830046,China)

机构地区：[1]新疆大学软件学院,新疆乌鲁木齐830091 [2]新疆大学信息科学与工程学院,新疆乌鲁木齐830046

出　　处：《中文信息学报》2019年第9期60-68,共9页Journal of Chinese Information Processing

基　　金：国家自然科学基金(61563051,61662074,61262064);国家自然科学基金(61331011);新疆自治区科技人才培养项目(QN2016YX0051);新疆天山青年计划项目(2017Q011)

摘　　要：该文提出一种基于注意力机制(attention mechanism,ATT)、独立循环神经网络(independently recurrent neural network,IndRNN)和卷积神经网络(convolutional neural network,CNN)结合的维吾尔语名词指代消解模型(ATT-IndRNN-CNN)。根据维吾尔语的语法和语义结构,提取17种规则和语义信息特征。利用注意力机制作为模型特征的选择组件计算特征与消解结果的关联度,结果分别输入IndRNN和CNN得到包含上下文信息的全局特征和局部特征,最后融合两类特征并使用softmax进行分类完成消解任务。实验结果表明,该方法优于传统模型,准确率为87.23%,召回率为88.80%,F值为88.04%,由此证明了该模型的有效性。This paper proposes an Uyghur nouns anaphora resolution model ATT-IndRNN-CNN based on Attention Mechanism(ATT),Independently Recurrent Neural Network(IndRNN)and Convolutional Neural Network(CNN).According to the grammar and semantic structure of Uyghur,17 rules and semantic information features are extracted.The attention mechanism is applied to select the features via the correlation between the features and the resolution results.The results are input into IndRNN and CNN to obtain the global features and local features in the context,respectively.Finally,the two types of features are merged and softmax is used to classify the resolution task.The experimental results show that the proposed method is better than the classical models,achieving the precision of 87.23%,the recall of 88.80%,and the F-measure of 88.04%.

关 键 词：注意力机制 独立循环神经网络 CNN 指代消解 维吾尔语 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]