结合词性规则和依存句法分析的评价对象抽取方法

doi:10.3969/j.issn.1006-2475.2016.04.004

计算机与现代化 ›› 2016, Vol. 0 ›› Issue (4): 16-20.doi: 10.3969/j.issn.1006-2475.2016.04.004

结合词性规则和依存句法分析的评价对象抽取方法

（1.广西财经学院实验教学中心，广西南宁530003;2.国海证券股份有限公司，广西南宁530028)

收稿日期:2016-01-19 出版日期:2016-04-14 发布日期:2018-09-30
作者简介:张建华(1986-)，女，河南辉县人，广西财经学院实验教学中心助理工程师，硕士，研究方向：文本处理，数据挖掘；肖中正（1988-），男，广西桂林人，国海证券股份有限公司助理工程师，硕士，研究方向：分布式计算，数据分析。

Opinion Target Extraction Method with Speech Rules and Dependency Parsing

(1. The Experimental Teaching Center, Guangxi University of Finance and Economics, Nanning 530003, China;2. Sealand Securities Co. Ltd., Nanning 530028, China)

Received:2016-01-19 Online:2016-04-14 Published:2018-09-30

摘要/Abstract

摘要：

评价对象的抽取能够让用户和商家同时受益，商家通过评价对象了解用户关心的产品特征，改进商品质量；用户通过评价对象做出购买决策。由于网络评论环境特殊，评价对象的抽取比传统的信息处理更复杂。在一些学者研究的基础上，本文提出一种词性规则和依存句法分析相结合的抽取方法。首先，该抽取方法利用词性规则制定名词短语抽取模板，得到候选评价对象，根据评价词对评价对象的修饰作用对评价对象进行第一次筛选；其次，利用8种依存句法关系对评价对象进行第二次筛选；最后，将2种筛选结果进行结合，得到最终的评价对象。实验结果表明，该方法在3类数据集上都取得了一定的效果。

关键词:

Abstract:

Abstract: Extracting opinion targets simultaneously benefit users and businesses, as businesses understand the product features by the opinion target of interest to users, then improve product quality, and users can make purchasing decisions by these objects. Due to the special circumstances of Web reviews, the extraction of opinion targets is more complex than conventional information processing. Based on some scholars’ basic researches, the paper proposed an extracting method for speech rules and dependency parsing. Firstly, speech rule is used to make noun phrase extraction template, getting candidate opinion targets that are filtered for the first time according to the modification of evaluation terms to opinion target. Secondly, eight kinds of dependency relationship are used to filter opinion targets for the second time. Finally, two kinds of screening results are combined to give the final opinion targets. Experimental results demonstrate that this method on three types of data sets has achieved certain results.

Key words: opinion target, evaluation term, information extraction, speech rules, dependency parsing

中图分类号:

TP391

张建华1,肖中正2.

结合词性规则和依存句法分析的评价对象抽取方法[J]. 计算机与现代化, 2016, 0(4): 16-20.

ZHANG Jian-hua1, XIAO Zhong-zheng2.

Opinion Target Extraction Method with Speech Rules and Dependency Parsing[J]. Computer and Modernization, 2016, 0(4): 16-20.

参考文献

［1］宋晓雷,王素格,李红霞. 面向特定领域的产品评价对象自动识别研究［J］. 中文信息学报, 2010,24(1):89-93.

［2］刘鸿宇,赵妍妍,秦兵,等. 评价对象抽取及其倾向性分析［J］. 中文信息学报, 2010,24(1):84-88.

［3］赵妍妍,秦兵,刘挺. 文本情感分析［J］. 软件学报, 2010,21(8):1834-1848.

［4］戴敏,王荣洋,李寿山,等. 基于句法特征的评价对象抽取方法研究［J］. 中文信息学报, 2014,28(4):92-97.

［5］ Hu Minqing, Liu Bing. Mining and summarizing customer reviews［C］// Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2004:168-177.

［6］ Wei Jin, Ho Hay Hung. A novel lexicalized HMM-based learning framework for Web opinion mining［C］// Proceedings of the 26th Annual International Conference on Machine Learning. 2009:465-472.

［7］郑敏洁,雷志城,廖祥文,等. 基于层叠CRFs的中文句子评价对象抽取［J］. 中文信息学报, 2013,27(3):69-76.

［8］ Yu Jianxing, Zha ZhengJun, Wang Meng, et al. Domain-assisted product aspect hierarchy generation: Towards hierarchical organization of unstructured consumer reviews［C］// Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing. 2011:140-150.

［9］章剑锋,张奇,吴立德,等. 中文观点挖掘中的主观性关系抽取［J］. 中文信息学报, 2008,22(2):55-59.

［10］徐冰,赵铁军,王山雨,等. 基于浅层句法特征的评价对象抽取研究［J］. 自动化学报, 2011,37(10):1241-1247.

［11］Kessler J S, Nicolov N. Targeting sentiment expressions through supervised ranking of linguistic configurations［C］// Proceedings of the 3rd International Conference on Weblogs and Social Media. 2009:90-97.

［12］Qiu Guang, Liu Bing, Bu Jiajun, et al. Opinion word expansion and target extraction through double propagation［J］. Computational Linguistics, 2011,37(1):9-27.

［13］Titov I, McDonald R. Modeling online reviews with multi-grain topic models［C］// Proceedings of the 17th International Conference on World Wide Web. 2008:111-120.

［14］Brody S, Elhadad N. An unsupervised aspect-sentiment model for online reviews［C］// Proceedings of the 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics. 2010:804-812.

［15］哈工大社会计算与信息检索研究中心. 语言云(语言技术平台云LTP-Cloud)［EB/OL］. http://www.ltp-cloud.com/, 2015-08-24.

［16］赵军,黄昌宁. 基于转换的汉语基本名词短语识别模型［J］. 中文信息学报, 1999,13(2):1-7.

[1]	王正敏,张太红,李永可,白涛. FreeMarker模板引擎在线动态生成Excel和Word文档技术[J]. 计算机与现代化, 2016, 0(4): 109-113.
[2]	姬忠孝，江国华. 一种基于FTA和FDG的安全关键函数定位方法[J]. 计算机与现代化, 2016, 0(4): 85-89+122.
[3]	柳萌宇1，钟浩2，于海波1. 基于变更相似性的跨语言克隆检测方法[J]. 计算机与现代化, 2016, 0(4): 79-84+99.
[4]	朱亚会，陈丹，庄毅. 云数据中心基于能耗感知的虚拟机调度算法[J]. 计算机与现代化, 2016, 0(4): 74-78.
[5]	宋文灏1，钟浩2，于海波1. 一种有效的API搜索算法[J]. 计算机与现代化, 2016, 0(4): 59-64.
[6]	曹清云1,倪建军1,2,王康1,吴榴迎1. 一种改进的未知动态环境下机器人混合路径规划方法[J]. 计算机与现代化, 2016, 0(4): 54-58.
[7]	蔡亦清，张亚，张松. 基于MSP430F149的弹匣计数器设计[J]. 计算机与现代化, 2016, 0(4): 50-53.
[8]	翟师伟1，沈士根2,3，曹奇英1. 基于非合作博弈的无线Mesh网络接入控制[J]. 计算机与现代化, 2016, 0(4): 45-49+78.
[9]	马明阳. 基于可信硬件的智能手机短信加密方案[J]. 计算机与现代化, 2016, 0(4): 29-35.
[10]	王致强，乐嘉锦，陈德华. 基于甲亢临床指标的多维时间序列关联度分析[J]. 计算机与现代化, 2016, 0(4): 12-15.

结合词性规则和依存句法分析的评价对象抽取方法

Opinion Target Extraction Method with Speech Rules and Dependency Parsing

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 10

编辑推荐

Metrics

本文评价