基于多特征和深度神经网络的维吾尔文情感分类  

Uyghur sentiment classification based on multi-features and deep neural network

作　　者：买买提阿依甫 吾守尔·斯拉木[1] 艾斯卡尔·艾木都拉[1] 杨文忠[1] 帕丽旦·木合塔尔[1] Maimaitiayifu;Wushouer Silamu;Aisikaer Aimudoula;Yang Wenzhong;Palidan Muhetaer(College of Information Science&Engineering,Xinjiang University,Urumqi 830046,China)

机构地区：[1]新疆大学信息科学与工程学院,乌鲁木齐830046

出　　处：《计算机应用研究》2020年第5期1368-1374,1379,共8页Application Research of Computers

基　　金：国家自然科学基金资助项目(61363063);国家“973”重点基础研究计划基金资助项目(2014CB340506);新疆大学多语种重点实验室开放课题(XJDX0905-2013-01)。

摘　　要：针对传统机器学习的情感分类方法存在长距离依赖问题与深度学习存在忽略情感词库的弊端,提出了一种基于注意力机制与双向长短记忆网络和卷积神经网络模型相结合的维吾尔文情感分类方法。将多特征拼接向量作为双向长短记忆网络的输入来捕获文本上下文信息,使用注意力机制和卷积网络获取文本隐藏情感特征信息,有效增强了对文本情感语义的捕获能力。实验结果表明,该方法在二分类和五分类情感数据集上的F1值相比于机器学习方法分别提高了5.59%和7.73%。In order to solve the problem of long-distance dependence in traditional machine learning sentiment classification method and the disadvantage of ignoring the emotional lexicon in deep learning,this paper proposed a Uyghur sentiment classification method based on attention mechanism combined with bidirectional long-short term memory network and convolutional neural network model.It used the concatenated multi-feature vector as the input of the bidirectional long short-term memory network to capture the context information,and used the attention mechanism and convolution network to capture text hidden emotional feature information,which effectively enhanced the capture ability of the text sentiment semantics.The experimental results show that the F1 value of this method on two-category and five-category Uyghur sentiment data sets are higher than machine learning method 5.59%and 7.73%respectively.

关 键 词：情感分类 双向长短记忆网络 卷积神经网络 注意力机制 维吾尔语 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]