基於情緒分析對於社群網路上之諷刺的文本偵測

簡易檢索 / 詳目顯示

回結果列表

研究生：	亞紹克 KUMAR, MUPPALA-ASHOK
論文名稱：	基於情緒分析對於社群網路上之諷刺的文本偵測 Detection of Textual Sarcasm on Social Media through Emotion Analysis
指導教授：	陳宜欣 Chen, Yi-Shin
口試委員:	陳朝欽 Chen, Chour-Chin 張隆紋 Chang, Long-wen
學位類別：	碩士 Master
系所名稱：
論文出版年：	2017
畢業學年度：	105
語文別：	英文
論文頁數：	39
中文關鍵詞：	推特、諷刺、網路欺凌、情感分析
外文關鍵詞：	Twitter, Sarcasm, Cyberbullying, Emotion Analysis
相關次數：	點閱：1 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

近年來社群網路的使用已大為普及,如社群網站Twitter已成為一個相當普遍的網路媒介令使用者得以透過簡單的Tweet分享他們的日常生活與意見。但卻有人正利用網路平台,對準他人以諷刺的形式進行羞辱,然而諷刺雖是不該出現的攻擊行為,卻難以透過傳統方法徵測出,有許多透過字彙及語言學的方法,對於產品的貧論或對Twitter進行諷刺的方法已被研究,但本論文著眼於不同的方法,且由於我們找出了相似的字詞特徵,此方法可用於諷刺及網路霸凌的徵測。(一般而言霸凌是以不直接的方式傷人,而網路霸凌則是用直接的形式傷人。)因此我們提出了一個新穎的方法,利用情緒分析以徵測諷刺與網路霸凌。盡我們所知,非常少的研究者曾使用情緒分析去徵測諷刺以及網路霸凌。因此我們使用情緒分析從Twitter中取出情緒相關的特徵,然後利用這些特徵來協助我們的分類器,我們的研究結果顯示出獲得了比現有的方法還要好的顯示成果。

With the popularity of social networks such as, Twitter became a popular broadcast medium
in recent years for users to share their daily activities and opinions by posting a simple
tweet. But some people are taking advantage of this platform in the form of targeting or
insulting someone in sarcastic manner. However, sarcasm is summarized as unwanted act
of aggression and it is difficult to detect through traditional means. Several researches
have been done on detecting of sarcasm on product reviews and twitter by using lexical
and linguistic cues. But this thesis looks at different methods that can be used for both
sarcasm and cyberbullying detection. Because we find out that there are some similar
characteristics between sarcasm and cyberbullying (sarcasm is an indirect form of hurting,
while cyberbullying is direct form of hurting).
So we come up with a new approach to detect both sarcasm and cyberbullying using
emotion analysis. As best of our knowledge very few researchers have done using emotion
analysis to detect sarcasm and cyberbullying. So, we perform emotion analysis to extract
emotion related features from the tweets. And using these features for our classification.
Our experiment results shows that better performance than the existing methods.
Keywords: Twitter, Sarcasm, Cyberbullying, Emotion analysis.

摘要 i
Abstract ii
Acknowledgement iii
List of Tables vii
List of Figures viii
Introduction 1
1 Sarcasm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Problem Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Related Work 7
1 Social Media Efforts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Emotion Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3 Sarcasm Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4 Cyberbullying Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Overview 10
Methodology 12
1 Data Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2 Feature Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Feature Type 1: Emotion of the Tweet (EMO) . . . . . . . . . . . . 14
2.2 Feature Type 2: Emotion score of each emotion for each tweet (ESEMO)
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3 Feature Type 3: Emotion score ranges of consecutive emotions
(ES-RANGE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4 Derived Features . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.5 Features for Classification . . . . . . . . . . . . . . . . . . . . . . 15
3 Feature Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5 Emotion Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.1 Emotion Classifier . . . . . . . . . . . . . . . . . . . . . . . . . . 18
5.2 Pattern Ranking . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Experimental Setup 21
1 System Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2 Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.1 Selecting a suitable learning algorithm . . . . . . . . . . . . . . . . 24
3 Baseline setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
v
3.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4 Main Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
5 Experiment Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Conclusion and Future Work 36
1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2 Future Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
References 38
                                

[1] Robert Plutchik. The emotions. University Press of America, 1991.
[2] Carlos Argueta, Elvis Saravia, and Yi-Shin Chen. Unsupervised graph-based patterns
extraction for emotion classification. In 2015 IEEE/ACM International Conference
on Advances in Social Networks Analysis and Mining (ASONAM), pages 336–341.
IEEE, 2015.
[3] Marc A Brackett, Susan E Rivers, Sara Shiffman, Nicole Lerner, and Peter Salovey.
Relating emotional abilities to social functioning: a comparison of self-report and
performance measures of emotional intelligence. Journal of personality and social
psychology, 91(4):780, 2006.
[4] Dmitry Davidov, Oren Tsur, and Ari Rappoport. Semi-supervised recognition of sarcastic
sentences in twitter and amazon. In Proceedings of the fourteenth conference
on computational natural language learning, pages 107–116. Association for Computational
Linguistics, 2010.
[5] Ellen Riloff, Ashequl Qadir, Prafulla Surve, Lalindra De Silva, Nathan Gilbert, and
Ruihong Huang. Sarcasm as contrast between a positive sentiment and negative situation.
In EMNLP, volume 13, pages 704–714, 2013.
[6] Roberto Gonz´alez-Ib´anez, Smaranda Muresan, and Nina Wacholder. Identifying sarcasm
in twitter: a closer look. In Proceedings of the 49th Annual Meeting of the
Association for Computational Linguistics: Human Language Technologies: short
papers-Volume 2, pages 581–586. Association for Computational Linguistics, 2011.
[7] Santosh Kumar Bharti, Korra Sathya Babu, and Sanjay Kumar Jena. Parsing-based
sarcasm sentiment recognition in twitter data. In 2015 IEEE/ACM International
Conference on Advances in Social Networks Analysis and Mining (ASONAM), pages
1373–1380. IEEE, 2015.
[8] Kelly Reynolds. April kontosthatis and lynne edwards using machine learning to
detect cyberbullying. NationalScience Foundation, Grant, (0916152):1–4, 2011.
[9] Karthik Dinakar, Roi Reichart, and Henry Lieberman. Modeling the detection of
textual cyberbullying. The Social Mobile Web, 11:02, 2011.
[10] Ellen Spertus. Smokey: Automatic recognition of hostile messages. In AAAI/IAAI,
pages 1058–1065, 1997.

簡易檢索 / 詳目顯示

相關論文