面向热门话题的微博观点挖掘方法

doi:10.3969/j.issn.1000-436x.2014.z2.030

Abstract

Abstract:

A method of opinion mining was argued in connection with hot topic of micro-blog.Firstly,opinion sentences are identified by syntactic dependencies templates and support vector machine (SVM).Secondly,the opinion words can be get in opinion sentences by syntactic dependencies and lexical relations,thus they can be displayed clearly.Finally,the experiment proves the effectiveness of this method.

Key words: micro-blog, hot topic, opinion sentences, opinion mining, syntactic dependencies

Guang-lei ZHANG,Ya-bin XU. Opinion mining method on hot topic of micro−blog[J]. Journal on Communications, 2014, 35(Z2): 220-227.

Figures/Tables 8

References 11

[1]	WIEBEM J , et al . Development and use of a gold standard data set for subjectivity classifications[A]. Proc 37thAnnual Meeting of the Association,for Computational Linguistics(ACL-99)[C]. NJ:Association for Computational Linguistics, 1999. 246-253
[2]	叶强, 张紫琼, 罗振雄 . 面向互联网评论情感分析的中文主观性自动判别方法[J]. 信息系统学报, 2007,1(1): 79-91. YE Q , ZHANG Z Q , LUO Z X . Automatically measuring subjectivity of Chinese sentence for sentiment analysis to reviews on Internet[J]. China Journal of Information Systems, 2007,1(1): 79-97.
[3]	蒙新泛, 王厚峰 . 主客观识别中的上下文因素的研究[A]. Advances of Computational Linguistics in China(2007-2009)[C]. 山东:中国中文信息学会, 2009. 594-599. MENG X F , WANG H F . A study on the impact of context information in subjectivity detection[A]. Advances of Computational Linguistics in China(2007-2009)[C]. Shandong:CIPS, 2009. 594-599.
[4]	PANG B , et al . A sentimental education:Sentiment analysis using subjectivity summarization based on minimum cuts[A]. Proceedings of the association for Computational Linguistics(ACL)[C]. PA:Association for Computational Linguistics, 2004. 271-278.
[5]	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. ZHANG J F , ZHANG Q , WU L D ,et al. Subjective relation extraction in Chinese opinion mining[J]. Journal of Chinese information, 2008,22(2): 55-59.
[6]	张博 . 基于 SVM 的中文观点句抽取[D]. 北京:北京邮电大学, 2011. ZHANG B . Chinese Opinion Sentence Extraction Based on SVM Classification[D]. Bejing:Beijing University of Posts and Telecommunications, 2011.
[7]	邓乃扬, 田英杰 . 支持向量机:理论,算法与拓展[M]. 北京: 科学出版社, 2009. 45-50. DENG N Y , TIAN Y J . Support Vector Machines:Theory,Algorithm and Development[M]. Beijing: Science PressPress, 2009. 45-50.
[8]	刘群, 李素建 . 基于《知网》的词汇语义相似度计算[A]. 第3届中文词汇语义学研讨会论文集[C]. 2002. 59-76. LIU Q , LI S J . Word similarity computing based on HowNet[A]. 3th ISCID[C]. 2002. 59-76.
[9]	NLP＆CC2012.中文微博情感分析测评[EB/OL]. .
[10]	DONG Z D , DONG Q . HowNet[EB/OL]. .
[11]	KU L W ,et al. NTUSD[EB/OL]. .

Metrics

Recommended 0

No Suggested Reading articles found!

关系类型	标记
主谓关系	SBV
动宾关系	VOB
状中结构	ADV
动补结构	CMP
核心词	HED

模板	观点句中出现次数	非观点句中出现次数
SBV认为VOB	48	1
SBV表示VOB	43	4
ADV就是VOB	42	6
ADV认为VOB	33	2
SBV觉得VOB	30	1

特征编号	特征	取值
1	是否含有情感词	0,1
2	是否含有主张词	0,1
3	是否含有程度副词	0,1
4	是否含有问号	0,1
5	是否含有感叹号	0,1
6	句法依存关系的个数	num
7	连词个数	num
8	是否含有表情符号	0,1

词法关系	提取方式	示例
n/ws+程度副词+adj	＜adj，n＞	例如：图像/n 非常/d 细腻/a
n/ws+程度副词+adj	＜adj，n＞	提取：＜图像，细腻＞
n/ws+adj	＜adj，n＞	例如：方舟子/n 无聊/a
n/ws+adj	＜adj，n＞	提取：＜方舟子，无聊＞
adj+的+n/ws	＜adj，n＞	例如：很好/adj 的学生/n
adj+的+n/ws	＜adj，n＞	提取：＜学生，很好＞

热门话题	观点句	合并前的观点词对	合并后的观点词对	出现频次
	1.大葱贵！	＜大葱，贵＞	＜大葱，贵＞	127
	2.葱贵死了！	＜葱，贵＞	＜大葱，贵＞	127
#疯狂的大葱#	3.季节的因素很关键，关键是工资太低！	＜工资，低＞	＜工资,低＞	98
	4.这年头物价一个劲儿涨，就是工资不涨。	＜物价，涨＞	＜物价，涨＞	84
	5.物价上涨的厉害啊！	＜物价，上涨＞	＜物价，涨＞	84

Opinion mining method on hot topic of micro−blog

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 8

References 11

Related Articles 5

Metrics

Recommended 0

[1]	. Opinion mining method on hot topic of micro?blog [J]. Journal on Communications, 2014, 35(Z2): 30-227.
[2]	. Model to find incarnated accounts in micro-blogging for public opinion supervision [J]. Journal on Communications, 2014, 35(Z2): 31-232.
[3]	Jun-li BU,Can PENG,Yi ZHENG,Jiu-ming HUANG,Bin ZHOU. Model to find incarnated accounts in micro-blogging for public opinion supervision [J]. Journal on Communications, 2014, 35(Z2): 228-232.
[4]	Min HE,Li-hong WANG,Pan DU,Jin ZHANG,Xue-qi CHENG. Microblog hot topic detection method based on meaningful string clustering [J]. Journal on Communications, 2013, 34(Z1): 256-262.
[5]	. Microblog hot topic detection method based on meaningful string clustering [J]. Journal on Communications, 2013, 34(Z1): 34-262.