ABSTRACT
In this paper, we extensively evaluate the effectiveness of using a user's social media activities for estimating degree of depression. As ground truth data, we use the results of a web-based questionnaire for measuring degree of depression of Twitter users. We extract several features from the activity histories of Twitter users. By leveraging these features, we construct models for estimating the presence of active depression. Through experiments, we show that (1) features obtained from user activities can be used to predict depression of users with an accuracy of 69%, (2) topics of tweets estimated with a topic model are useful features, (3) approximately two months of observation data are necessary for recognizing depression, and longer observation periods do not contribute to improving the accuracy of estimation for current depression; sometimes, longer periods worsen the accuracy.
- Barry, D. T. Cultural and demographic correlates of self-reported guardedness among East Asian immigrants in the US. International Journal of Psychology 38, 3 (2003), 150--159.Google ScholarCross Ref
- Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., and Erbaugh, J. An inventory for measuring depression. Archives of General Psychiatry 4, 6 (June 1961), 561--571.Google ScholarCross Ref
- Bengio, Y. Learning deep architectures for AI. Foundations and Trends in Machine Learning 2, 1 (Jan. 2009), 1--127. Google ScholarDigital Library
- Bengio, Y., Courville, A., and Vincent, P. Representation learning: A review and new perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence 35, 8 (Aug. 2013), 1798--1828. Google ScholarDigital Library
- Blei, D. M., Ng, A. Y., and Jordan, M. I. Latent dirichlet allocation. The Journal of Machine Learning Research 3 (Jan. 2003), 993--1022. Google ScholarDigital Library
- Bollen, J., Mao, H., and Zeng, X. Twitter mood predicts the stock market. Journal of Computational Science 2, 1 (Mar. 2011), 1--8.Google ScholarCross Ref
- Bradley, M., Lang, P., and Cuthbert, B. Affective norms for english words (ANEW). Gainsville, Fla, NIMH Centre for the Study of Emotion and Attention, University of Florida (1999).Google Scholar
- Breiman, L. Bagging predictors. Machine Learning 24, 2 (Aug. 1996), 123--140. Google ScholarDigital Library
- Breiman, L. Random forests. Machine Learning 45, 1 (Oct. 2001), 5--32. Google ScholarDigital Library
- Culotta, A. Estimating county health statistics with twitter. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI'14) (Apr. 2014), 1335--1344. Google ScholarDigital Library
- De Choudhury, M., Counts, S., and Horvitz, E. Predicting postpartum changes in emotion and behavior via social media. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI'13) (Apr. 2013), 3267--3276. Google ScholarDigital Library
- De Choudhury, M., Counts, S., and Horvitz, E. Social media as a measurement tool of depression in populations. In Proceedings of the 5th Annual ACM Web Science Conference (WebSci'13) (May 2013), 47--56. Google ScholarDigital Library
- De Choudhury, M., Counts, S., Horvitz, E. J., and Hoff, A. Characterizing and predicting postpartum depression from shared facebook data. In Proceedings of the 17th ACM Conference on Computer Supported Cooperative Work & Social Computing (CSCW'14) (Feb. 2014), 626--638. Google ScholarDigital Library
- De Choudhury, M., Gamon, M., Counts, S., and Horvitz, E. Predicting depression via social media. In Proceedings of the 7th International AAAI Conference on Weblogs and Social Media (ICWSM'13) (July 2013), 128--137.Google Scholar
- Government of Japan. White paper on goverment measures for suicide, June 2014. Also available as http://www8.cao.go.jp/jisatsutaisaku/ whitepaper/index-w.html (in Japanese).Google Scholar
- Griffiths, T. L., and Steyvers, M. Finding scientific topics. Proceedings of the National Academy of Sciences of the USA of America 101, Suppl 1 (Apr. 2004), 5228--5235.Google Scholar
- Hamilton, M. Development of a rating scale for primary depressive illness. British Journal of Social & Clinical Psychology 6, 4 (Dec. 1967), 278--296.Google ScholarCross Ref
- Houston, T. K., Cooper, L. A., Vu, H. T., Kahn, J., Toser, J., and Ford, D. E. Screening the public for depression through the Internet. Psychiatric Services 52, 3 (Mar. 2001), 362--367.Google ScholarCross Ref
- Inui, T., Umezawa, Y., and Yamamoto, M. Complaint sentence detection via automatic training data generation using sentiment lexicons and context coherence. Journal of Natural Language Processing 20, 5 (Dec. 2013), 683--705. (in Japanese).Google ScholarCross Ref
- Kudo, T. Mecab: Yet another part-of-speech and morphological analyzer. http://mecab.sourceforge.net/.Google Scholar
- Marcus, M., Yasamy, M. T., van Ommeren, M., Chisholm, D., and Saxena, S. Depression: A global public health concern. Tech. rep., WHO Dept. of Mental Health and Substance Abuse, Oct. 2012. Also available as http://www.who.int/mental_health/management/ depression/who_paper_depression_wfmh_2012.pdf.Google Scholar
- Moreno, M. A., Jelenchick, L. A., Egan, K. G., Cox, E., Young, H., Gannon, K. E., and Becker, T. Feeling bad on Facebook: Depression disclosures by college students on a social networking site. Depression and Anxiety 28, 6 (June 2011), 447--455.Google ScholarCross Ref
- O'Connor, B., Balasubramanyan, R., Routledge, B. R., and Smith, N. A. From tweets to polls: Linking text sentiment to public opinion time series. In Proceedings of the 4th International AAAI Conference on Weblogs and Social Media (ICWSM'10) (May 2010), 122--129.Google Scholar
- OECD. Health at a glance 2013, Nov. 2013.Google Scholar
- Ono, T., and Thompson, S. A. Japanese (w)atashi/ore/boku I: They're not just pronouns. Cognitive Linguistics 14, 4 (2003), 321--348.Google ScholarCross Ref
- Pak, A., and Paroubek, P. Twitter as a corpus for sentiment analysis and opinion mining. In Proceedings of the 7th International Conference on Language Resources and Evaluation (LREC'10) (May 2010), 1320--1326.Google Scholar
- Park, M., Cha, C., and Cha, M. Depressive moods of users portrayed in twitter. In Proceedings of the ACM SIGKDD Workshop on Healthcare Informatics (HI-KDD'12) (Aug. 2012), 1--8.Google Scholar
- Park, S., Lee, S. W., Kwak, J., Cha, M., and Jeong, B. Activities on Facebook reveal the depressive state of users. Journal of Medical Internet Research 15, 10 (Oct. 2013), e217.Google ScholarCross Ref
- Paul, M. J., and Dredze, M. You are what you tweet: Analyzing twitter for public health. In Proceedings of the 5th International AAAI Conference on Weblogs and Social Media (ICWSM'11) (July 2011), 265--272.Google Scholar
- Radloff, L. S. The CES-D scale a self-report depression scale for research in the general population. Applied Psychological Measurement 1, 3 (1977), 385--401.Google ScholarCross Ref
- Resnik, P., Garron, A., and Resnik, R. Using topic modeling to improve prediction of neuroticism and depression. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing (EMNLP '13) (Oct. 2013), 1348--1353.Google Scholar
- Rihmer, Z. Can better recognition and treatment of depression reduce suicide rates? A brief review. European Psychiatry 16, 7 (Nov. 2001), 406--409.Google ScholarCross Ref
- Si, J., Mukherjee, A., Liu, B., Li, Q., Li, H., and Deng, X. Exploiting topic based twitter sentiment for stock prediction. In Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (ACL'13) (Aug. 2013), 24--29.Google Scholar
- Simon, G., Goldberg, S., Tiemens, B., and Ustun, T. Outcomes of recognized and unrecognized depression in an international primary care study. General Hospital Psychiatry 21, 2 (Mar. 1999), 97--105.Google ScholarCross Ref
- Takamura, D., Inui, T., and Okumura, M. Extracting semantic orientations using spin model. IPSJ Journal 47, 2 (Feb. 2006), 627--637. (in Japanese).Google Scholar
- Tiemens, B., Ormel, J., Jenner, J., Van der Meer, K., Van Os, T., Van Den Brink, R., Smit, A., and Van den Brink, W. Training primary-care physicians to recognize, diagnose and manage depression: does it improve patient outcomes? Psychological Medicine 29, 04 (July 1999), 833--845.Google ScholarCross Ref
- Tsugawa, S., Mogi, Y., Kikuchi, Y., Kishino, F., Fujita, K., Itoh, Y., and Ohsaki, H. On estimating depressive tendency of twitter users from their tweet data. In Proceedings of the 2nd International Workshop on Ambient Information Technologies (AMBIT'12) (Mar. 2013), 29--32.Google Scholar
- Tumasjan, A., Sprenger, T., Sandner, P., and Welpe, I. Predicting elections with twitter: What 140 characters reveal about political sentiment. In Proceedings of the 4th International AAAI Conference on Weblogs and Social Media (ICWSM'10) (May 2010), 178--185.Google Scholar
- Zhang, X., Fuehres, H., and Gloor, P. Predicting stock market indicators through twitter? I hope it is not as bad as I fear. Procedia-Social and Behavioral Sciences 26 (2011), 55--62.Google ScholarCross Ref
- Zhung, W. W. K. A self-rating depression scale. Archives of General Psychiatry 12, 1 (1965), 63--70.Google Scholar
Index Terms
- Recognizing Depression from Twitter Activity
Recommendations
Social media as a measurement tool of depression in populations
WebSci '13: Proceedings of the 5th Annual ACM Web Science ConferenceDepression is a serious and widespread public health challenge. We examine the potential for leveraging social media postings as a new type of lens in understanding depression in populations. Information gleaned from social media bears potential to ...
A content analysis of depression-related tweets
This study examines depression-related chatter on Twitter to glean insight into social networking about mental health. We assessed themes of a random sample (n = 2000) of depression-related tweets (sent 4-11 to 5-4-14). Tweets were coded for expression ...
Facebook use, envy, and depression among college students
A survey of 736 college students found that Facebook use can trigger feelings of envy.Feelings of envy were found to predict depression symptoms.The effect of surveillance use of Facebook on depression is mediated by feelings of envy.Surveillance use of ...
Comments