躁鬱症是影響全球約6000萬人的幾種嚴重情緒障礙之一。由於混亂的複雜性，檢測仍然是一項艱鉅的任務。為了有助於研究社會媒體中的躁鬱症患者，一種方法被設計為在社會媒體的個人層面自動檢測抑鬱和輕躁狂事件，這些診斷時間段和行為標準在“精神障礙診斷與統計手冊”（DSM-V）躁鬱症障礙標準。量化社交媒體環境中的躁鬱症行為;情緒，情感和語言特徵被使用。此外，提出了一種於網絡可視化的工具來探索關於參數的躁鬱症行為。幾個標準達到70％以上的精準度。結果表明，標籤和抑鬱症狀的詞語可以作為躁鬱症行為的其他指標。發現躁鬱症患者在發作期間比其他時間更頻繁地提及宗教靈感的詞彙。發現躁鬱症用戶和正常用戶之間呈現出抑鬱行為的明顯差異。這種方法可以改善診斷技術，提供可以臨床心理學家和患者的及時分析結果。此外，該方法可檢測出其他重大事件，如疾病爆發前。

Bipolar disorder is one of several severe emotional disorders that affect approximately 60 million people worldwide. Due to the complexities of the disorder, it remains a difficult task to detect. To contribute to the study of bipolar disorder in social media a method is designed to auto detect depressive and hypomanic episodes at an individual level in social media with respect to diagnostic time periods and behavioural criteria that is defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) criteria for bipolar disorder. To quantify bipolar behaviours in a social media setting; sentiment, emotion and linguistic features are employed. Furthermore, a web based visualization tool is proposed for exploring bipolar behaviours with respect to spatio-temporal parameters. Several criterias achieved an accuracy above 70%. Furthermore, results suggest that awareness hashtags and depressive symptoms keywords in particular can serve as additional indicators of depressive behaviour in social media. It was found that bipolar patients refer to religious inspiration keywords more often than any other keywords during hypomanic episodes. Emotion features were found to present clear differences in depressive behaviour between bipolar and normal users. This method can improve diagnosis techniques and provide prompt analysis services and results which can support clinical psychologists and patients. In addition, this method can be customized to detect other major events in real-time such as disease outbreaks.

摘要 i
Abstract ii
Acknowledgement iii
List of Figures v
List of Tables vi
1 Introduction 1
2 Related Works 5
3 Methodology 8
3.1 Data Collection 9
3.2 Data Pre-processing and Cleaning 11
3.3 Data Preparation 13
3.3.1 Sentiment labelling 13
3.3.2 Mood labelling 13
3.3.3 Keyword labelling 16
4 Experiments and Discussion 19
4.1 Exploratory Analysis 19
4.2 Language Use 24
4.3 Auto-detection of Bipolar Behaviours: A Temporal Approach 25
4.3.1 Auto Detection of Bipolar Behaviours Using Sentiment and Linguistic Features 27
4.3.2 Auto Detection of Bipolar Behaviours Using Emotion and Linguistic Features 30
4.4 Auto-detection of Bipolar Behaviours: A Spatio-Temporal Approach 33
4.4.1 Auto Detection of Bipolar Behaviours Using Sentiment and Linguistic Features 34
4.4.2 Visualization Tool for exploring bipolar behaviours 37
5 Conclusion 38
References 40

[1] World Health Organization. Mental Disorders Fact Sheet. 2017. http://www.who.int/mediacentre/factsheets/fs396/en/
[2] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders DSM-V. 5th Edition. 2013. ISBN: 978-0-89042-554-1.
[3] Francis Mark Mondimore. Bipolar Disorder: A Guide for Patients and Families. 2nd ed. Johns Hopkins University Press. 2006. https://ebookcentral-proquest-com.nthulib-oc.nthu.edu.tw/lib/nthutw/detail.action?docID=4642475
[4] Michael J. Paul and Mark Dredze. You are what you tweet: Analysing Twitter for public health. International AAAI Conference on Web and Social Media, 20:265-272, 2011.
[5] C. Mair, A. V. Roux Diez, T. L. Osypuk, S. R. Rapp, T. Seeman and K. E. Watson. Is neighbourhood racial/ethnic composition associated with depressive symptoms? Social Science and Medicine. 71(3):541-550, 2010.
[6] Wei Yang and Lan Mu. GIS analysis of depression among Twitter users. Applied Geography, 60:217-223, 2015.
[7] Glen Coppersmith, Mark Dredze and Craig Harman. Quantifying Mental Health Signals in Twitter. In Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality. Association for Computational Linguistics, pages 51-60, Baltimore, Maryland USA, 2014.
[8] Elvis Saravia, Chun-Hao Chang, Renaud Jollet De Lorenzo, Yi-Shin Chen. MIDAS: Mental Illness Detection in Social Media. In Proceedings of the 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pages 1418-1421. San Francisco, CA, USA. IEEE, 2016.
[9] Glen Coppersmith, Mark Dredze, Craig Harman and Kristy Hollingshead. From ADHD to SAD: Analyzing the Language of Mental Health on Twitter through Self-Reported Diagnoses. In Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality. Association for Computational Linguistics, pages 1-10, Denver, Colorado, 2015.
[10] Mike Conway and Daniel O’Connor. Social media, big data, and mental health: current advances and ethical implications. Current Opinion in Psychology 2016, 9:77-82, 2016.
[11] Munmun De Choudhury, Scott Counts and Eric Horvitz. Social Media as a Measurement Tool of Depression in Populations. In Proceedings the 5th Annual ACM Web Science Conference, pages 47-56, Paris, France, ACM, 2013.
[12] Munmun De Choudhury, Michael Gamon, Scott Counts and Eric Horvitz. Predicting Depression via Social Media. In Proceedings of the 7th International AAAI Conference on Weblogs and Social Media 2013, pages 128-137. AAAI Press, 2013.

[13] Adam D. I. Kramer, Susan R. Fussell, Leslie D. Setlock. Text Analysis as a Tool for Analyzing Conversation in Online Support Groups. In Proceedings of CHI 2004, pages 1485-1488. CHI, 2004.
[14] Xin Zheng, Jialong Han and Sun Aixin. A Survey of Location Prediction on Twitter. arXiv:1705.03172. 2017.
[15] Jacopo Staiano and Marco Guerini. DepecheMood: a Lexicon for Emotion Analysis from Crowd-Annotated News. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics, pages 427-433, Baltimore, Maryland, USA. 2014.
[16] James A. Russell. A Circumplex Model of Affect. Journal of Personality and Social Psychology 1980, 39:1161-1178, 1980.
[17] Bridianne O'Dea, Stephen Wan, Philip J. Batterham, Alison L. Calear, Cecile Paris, Helen Christensen. Detecting suicidality on Twitter. Internet Interventions, 2(2):183-188, 2015.
[18] Minsu Park, Chiyoung Cha, and Meeyoung Cha. Depressive moods of users portrayed in twitter. In Proceedings of the CM SIGKDD Workshop on Healthcare Informatics, HI-KDD, pages 1-8, Beijing, China. 2012.
[19] Xiang Ji, Soon Ae Chun, Zhi Wei and James Geller. Twitter sentiment classification for measuring public health concerns. Social Network Analysis and Mining 2015, 5:13, 2015.
[20] Cheng and Thomas Wicks. Event Detection using Twitter: A Spatio-Temporal Approach. PLoS ONE, 9(6): e97807. doi:10.1371/journal.pone.0097807, 2014.
[21] Philip Resnik, William Armstrong, Leonardo Claudino, Thang Nguyen, Viet-An Nguyen and Jordan Boyd-Graber. Beyond LDA: Exploring Supervised Topic Modeling for Depression-Related Language in Twitter. In Proceedings of the 2nd Workshop on Computational Linguistics and Clinical Psychology. Association for Computational Linguistics, pages 99-107, Denver, Colorado. 2015.
[22] Arman Cohan, Sydney Young, Nazli Goharian. Triaging Mental Health Forum Posts. In Proceedings of the 3rd Workshop on Computational Linguistics and Clinical Psychology. Association for Computational Linguistics, pages 143-147, San Diego, California. 2016.
[23] Alec Go, Richa Bhayani and Lei Huang. Twitter Sentiment Classification using Distant Supervision. CS224N Project Report, Stanford, pages 1-12, 2009.
[24] Maryam Hasan, Elke Rundensteiner and Emmanuel Agu. EMOTEX: Detecting Emotions in Twitter Messages. In Proceedings of ASE BIGDATA/SOCIALCOM/CYBERSECURITY Conference, pages 1-10, Stanford University. 2014.
[25] Bo Pang, Lillian Lee, and Shivakumar Vaithyanathan. Thumbs up? Sentiment Classification using Machine Learning Techniques. In Proceedings of the Conference on Empirical Methods in Natural Language Processing. EMNLP 2002, pages 79-86, Philadelphia, 2002.
[26] Munmun De Choudhury, Scott Counts, and Michael Gamon. Not all moods are created equal! Exploring human emotional states in social media. In Proceedings of the 6th International Conference on Weblogs and Social Media (ICWSM'12). 2012