互聯網和永久性的增長在微博上的人氣普及
社交網絡在過去幾年中，也導致了一個顯著增加
用戶對這些網站微博上載的數據。但是所產生的數據是
動態的性質，綁時間條件和用戶的主體。
日常生活經驗，討論或事件對行為有直接影響
體現在社會網絡。因此，它是非常重要的驢
影響這些相互作用有超過一個社會群體。一種替代回答
這是確定如何影響一個主題是根據行為呈現
在社交網絡隨時間。然後，必須尋找和開發方法
可利用實現這一任務。這項工作結合相關的三個方面
這種數據的利用：話題識別，情感分類和影響力
決心。一旦主題確定我們首先它歸類為特定時間或
長期看，然後投遞相關的話題進行收集和每一個被分配了一個
情感標籤。處理崗位的流後有利於基於時間的分析
我們提出了基於這將給予每個主題的價值影響其得分
壽命，情感過渡，以量化的話題是如何影響力達到
在一個社會群體，特別是從檢測到的事件在twitter上。換句話說，我們
總結對事件的情緒反應，並結合其與時間
變量來確定它是在一個社會團體如何影響力。

The ubiquity of internet and permanent growth in popularity of Microblogging
Social Networks over the past years, has also led to a significant increase in the
data uploaded by users to these Microblog sites. However the generated data is
dynamic by nature, tied to temporal conditions and the subjectivity of its users.
Everyday life experiences, discussions or events have a direct impact on the behaviors
reflected in social networks. It is therefore of great importance to asses the
impact these interactions are having over a social group. An alternative to answer
this is determining how influential a topic is according to the behavior presented
on a social network over time. It is then necessary to find and develop methods
that can leverage towards this task. This work combines three fields relevant to
this kind of data utilization: topic identification, emotion classification and influence
determination. Once a topic is identified we first classify it as time specific or
long term, then posts relevant to the topic are collected and each one is assigned an
emotion label. After processing the stream of posts to favor a time based analysis
we propose an Influence Value score which will be given to each topic based on its
lifespan, emotion transition and reach in order to quantify how influential a topic is
over a social group, specifically from events detected on twitter. In other words we
summarize the emotional response towards an event and combine it with temporal
variables to determine how influential it is over a social group.

Contents
Introduction 1
Related Work 4
Methodology 7
3.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 Event Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.1 Pre-Processing . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.2.2 Keyword Extraction . . . . . . . . . . . . . . . . . . . . . . . 9
3.2.3 Event Graph Generation . . . . . . . . . . . . . . . . . . . . . 10
3.3 Emotion Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.3.1 Datasets and Pre-Processing . . . . . . . . . . . . . . . . . . . 12
3.3.2 Emotion Graph Construction . . . . . . . . . . . . . . . . . . . 13
3.3.3 Pattern Extraction . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3.4 The Classifier . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.4 Influence Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.4.1 Lifespan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.4.1.1 Outlier Detection for Temporal Data . . . . . . . . . 18
3.4.1.2 Time-Based Classification . . . . . . . . . . . . . . . 20
3.4.1.3 Topic Duration . . . . . . . . . . . . . . . . . . . . . 21
3.4.2 Emotion Transition . . . . . . . . . . . . . . . . . . . . . . . . 22
3.4.2.1 Emotion Distribution . . . . . . . . . . . . . . . . . 23
3.4.2.2 Transition Analysis . . . . . . . . . . . . . . . . . . 23
3.4.3 Reach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.4.4 The Score . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Experiments and Results 26
4.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.2 Precision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2.1 Emotion Classifier . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2.2 Collected Data . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.3 Emotion Distribution by Category . . . . . . . . . . . . . . . . . . . . 29
4.4 Transition by Emotion . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.5 Transition by Category . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.6 Influence Value Scores . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Conclusion 34

[1] Foteini Alvanaki, Michel Sebastian, Krithi Ramamritham, and Gerhard Weikum.
Enblogue: emergent topic detection in web 2.0 streams. In Proceedings of the
2011 ACM SIGMOD International Conference on Management of data, pages
1271–1274. ACM, 2011.
[2] Hila Becker, Mor Naaman, and Luis Gravano. Beyond trending topics: Real-world
event identification on twitter. ICWSM, 11:438–441, 2011.
[3] David M Blei and John D Lafferty. Dynamic topic models. In Proceedings of the
23rd international conference on Machine learning, pages 113–120. ACM, 2006.
[4] Giridhar Kumaran and James Allan. Text classification and named entities for
new event detection. In Proceedings of the 27th annual international ACM SIGIR
conference on Research and development in information retrieval, pages 297–304.
ACM, 2004.
[5] JianshuWeng and Bu-Sung Lee. Event detection in twitter. ICWSM, 11:401–408,
2011.
[6] George Valkanas and Dimitrios Gunopulos. How the live web feels about events.
In Proceedings of the 22nd ACM international conference on Conference on information
& knowledge management, pages 639–648. ACM, 2013.
[7] Hassan Sayyadi, Matthew Hurst, and Alexey Maykov. Event detection and tracking
in social streams. In ICWSM, 2009.
[8] Sasa Petrovic, Miles Osborne, Richard McCreadie, Craig Macdonald, and Iadh
Ounis. Can twitter replace newswire for breaking news? 2013.
[9] Qiming Diao, Jing Jiang, Feida Zhu, and Ee-Peng Lim. Finding bursty topics from
microblogs. In Proceedings of the 50th Annual Meeting of the Association for
35
Computational Linguistics: Long Papers-Volume 1, pages 536–544. Association
for Computational Linguistics, 2012.
[10] Liangjie Hong, Amr Ahmed, Siva Gurumurthy, Alexander J Smola, and Kostas
Tsioutsiouliklis. Discovering geographical topics in the twitter stream. In Proceedings
of the 21st international conference on World Wide Web, pages 769–778.
ACM, 2012.
[11] Manish Gupta, Jing Gao, Charu Aggarwal, and Jiawei Han. Outlier detection
for temporal data. Synthesis Lectures on Data Mining and Knowledge Discovery,
5(1):1–129, 2014.
[12] Bernard J Jansen, Mimi Zhang, Kate Sobel, and Abdur Chowdury. Twitter power:
Tweets as electronic word of mouth. Journal of the American society for information
science and technology, 60(11):2169–2188, 2009.
[13] Vipul Pandey and C Iyer. Sentiment analysis of microblogs. CS 229: Machine
learning final projects, 2009.
[14] Bo Pang and Lillian Lee. A sentimental education: Sentiment analysis using subjectivity
summarization based on minimum cuts. In Proceedings of the 42nd annual
meeting on Association for Computational Linguistics, page 271. Association
for Computational Linguistics, 2004.
[15] Prem Melville, Wojciech Gryc, and Richard D Lawrence. Sentiment analysis of
blogs by combining lexical knowledge with text classification. In Proceedings
of the 15th ACM SIGKDD international conference on Knowledge discovery and
data mining, pages 1275–1284. ACM, 2009.
[16] Bin Wei and Christopher Pal. Cross lingual adaptation: an experiment on sentiment
classifications. In Proceedings of the ACL 2010 Conference Short Papers,
pages 258–262. Association for Computational Linguistics, 2010.
[17] Alexander Pak and Patrick Paroubek. Twitter as a corpus for sentiment analysis
and opinion mining. In LREC, volume 10, pages 1320–1326, 2010.
[18] Dmitry Davidov, Oren Tsur, and Ari Rappoport. Enhanced sentiment learning
using twitter hashtags and smileys. In Proceedings of the 23rd International Conference
on Computational Linguistics: Posters, pages 241–249. Association for
Computational Linguistics, 2010.
[19] Kirk Roberts, Michael A Roach, Joseph Johnson, Josh Guthrie, and Sanda M
Harabagiu. Empatweet: Annotating and detecting emotions on twitter. In LREC,
pages 3806–3813, 2012.
[20] Yejin Choi and Claire Cardie. Learning with compositional semantics as structural
inference for subsentential sentiment analysis. In Proceedings of the Conference
on Empirical Methods in Natural Language Processing, pages 793–801. Association
for Computational Linguistics, 2008.
[21] Axel Bruns and Jean Burgess. Researching news discussion on twitter: New
methodologies. Journalism Studies, 13(5-6):801–814, 2012.
[22] Alison L Hill, David G Rand, Martin A Nowak, and Nicholas A Christakis. Emotions
as infectious diseases in a large social network: the sisa model. Proceedings
of the Royal Society of London B: Biological Sciences, 277(1701):3827–3835,
2010.
[23] Shaozhi Ye and FelixWu. Measuring message propagation and social influence on
twitter. com. International Journal of Communication Networks and Distributed
Systems, 11(1):59–76, 2013.
[24] Eytan Bakshy, Dean Eckles, Rong Yan, and Itamar Rosenn. Social influence in
social advertising: evidence from field experiments. In Proceedings of the 13th
ACM Conference on Electronic Commerce, pages 146–161. ACM, 2012.
[25] Lu Liu, Jie Tang, Jiawei Han, Meng Jiang, and Shiqiang Yang. Mining topic-level
influence in heterogeneous networks. In Proceedings of the 19th ACM international
conference on Information and knowledge management, pages 199–208.
ACM, 2010.
[26] Suin Kim, JinYeong Bak, and Alice Haeyun Oh. Do you feel what i feel? social
aspects of emotions in twitter conversations. In ICWSM, 2012.
[27] Allan Filipowicz, Sigal Barsade, and Shimul Melwani. Understanding emotional
transitions: the interpersonal consequences of changing emotions in negotiations.
Journal of personality and social psychology, 101(3):541, 2011.
[28] Rada Mihalcea and Paul Tarau. Textrank: Bringing order into texts. Association
for Computational Linguistics, 2004.
[29] Simon White. How to strike a match. Development Cycles, 2004.
38