Journal of the Korean Society for information Management (정보관리학회지)

Korean Society for Information Management (한국정보관리학회)
권호 표지
DOI QR코드

DOI QR Code

An Analysis of the Discourse Topics of Users who Exhibit Symptoms of Depression on Social Media

소셜미디어를 통한 우울 경향 이용자 담론 주제 분석

  • 서하림 (연세대학교 문헌정보학과) ;
  • 송민 (연세대학교 문헌정보학과)
  • Received : 2019.11.17
  • Accepted : 2019.12.19
  • Published : 2019.12.30
Download PDF
⟨ Previous Next ⟩

Abstract

Depression is a serious psychological disease that is expected to afflict an increasing number of people. And studies on depression have been conducted in the context of social media because social media is a platform through which users often frankly express their emotions and often reveal their mental states. In this study, large amounts of Korean text were collected and analyzed to determine whether such data could be used to detect depression in users. This study analyzed data collected from Twitter users who had and did not have depressive tendencies between January 2016 and February 2019. The data for each user was separately analyzed before and after the appearance of depressive tendencies to see how their expression changed. In this study the data were analyzed through co-occurrence word analysis, topic modeling, and sentiment analysis. This study's automated data collection method enabled analyses of data collected over a relatively long period of time. Also it compared the textual characteristics of users with depressive tendencies to those without depressive tendencies.

우울증은 전 세계적으로 많은 사람들이 겪고 있으며, 최근 다양한 분야에서 꾸준히 우울증에 대한 연구가 수행되고 있다. 특히 사람들이 본인의 스트레스나 감정 상태에 대해 소셜미디어에 공유한 글을 통해 그들의 심리나 정신건강에 대해 파악해보고자 하는 맥락에서 소셜미디어를 활용한 연구 역시 유의미하게 받아들여지고 있다. 이에 본 연구에서는 우울 경향의 이용자와 그렇지 않은 이용자들의 2016년부터 2019년 2월까지의 트위터 데이터를 수집하여 어떤 주제적, 어휘 사용의 특성을 보이는지 보고자 하였으며, 우울 경향의 시기별로도 어떤 차이를 보이는지 살펴보기 위해 우울 경향 관측 날짜를 기준으로 하여 이전(before) 시기와 이후(after) 시기를 구분하여 실험을 수행하였다. 토픽모델링, 동시출현 단어분석, 감성분석 방법을 통해 우울 경향과 비(非)우울 경향 이용자의 텍스트의 주제적 차이를 살펴보았고, 감성 반응에 따라 사용한 어휘에 대해서도 살펴봄으로써 어떠한 특성이 있는지 확인해 보았다. 데이터 수집 단계에서 '우울' 표현을 포함한 텍스트 데이터 수집방법을 통해 비교적 긴 기간, 많은 양의 데이터를 수집할 수 있었고, 또한 우울 경향의 여부와 시기적 구분에 따른 관심 주제에 대한 차이도 확인할 수 있었다는 점에서 유의미하다고 볼 수 있다.

Keywords

References

  1. Kim, Jae-bong, & Kim, Hyung-Joong (2017). A domain-specific sentiment lexicon construction method for stock index directionality. Digital Contents Society, 18(3), 585-592.
  2. Bae, Jung-hwan, Son, Ji-eun, & Song, Min (2013). Analysis of twitter for 2012 south Korea presidential election by text mining techniques. Journal of Intelligence and Information Systems, 19(3), 141-156. https://doi.org/10.13088/jiis.2013.19.3.141
  3. Seo, Sang-hyeon, & Kim, Jun-Tae (2016). Deep learning based sentiment analysis research trends. Korea Multimedia Society, 3, 8-22.
  4. Song, Min (2017). Text mining. Seoul: Chungram Books.
  5. Song, Ho-Yun, Park, Han-chul, Yang, Won-seok, & Park, Jong-chul (2017). Predicting symptoms of depression for social media users via linguistic patterns. Korea Information Science Society, 625-627.
  6. Yu, Eun-Ji, Kim, Yoo-sin, Kim, Nam-gyu, & Jeong, Seung-ryul (2013). Predicting the direction of the stock index by using a domain-specific sentiment dictionary. Journal of Intelligence and Information Systems, 19(1), 95-110. https://doi.org/10.13088/jiis.2013.19.1.095
  7. Lee, Sang Hoon, Jing Cui, & Kim, Jong-Woo (2016). Sentiment analysis on movie review through building modified sentiment dictionary by movie genre. Journal of Intelligence and Information Systems, 22(2), 97-113. https://doi.org/10.13088/jiis.2016.22.2.097
  8. Lee, Yu-Lim (2016). Pharmaceuticalization of emotion and structuring depression as experience: An analysis of the depression experiences of Korean women in their 20s. Korea Womens Studies Institute, 16(1), 81-117.
  9. Lee, Hyun-Seo, & Song, Min (2018). An analysis of the influences between sentiment values of korean online news and macroeconomic indicators using text mining. Journal of Communication Science, 18(3), 129-169. https://doi.org/10.14696/jcs.2018.09.18.3.129
  10. Chang, Jae-Young (2009). A sentiment analysis algorithm for automatic product reviews classification in on-line shopping mall. The Journal of Society for e-Business Studies, 14(4), 19-33.
  11. Jin, Seol-A, & Song, Min (2016). Topic modeling based interdisciplinarity measurement in the informatics related journals. Journal of the Korean Society for Information Management, 33(1), 7-32. https://doi.org/10.3743/kosim.2016.33.1.007
  12. Bastian, Heymann, & Jacomy (2009). Gephi: An open source software for exploring and manipulating networks. Icwsm, 361-362.
  13. Coppersmith, G., Dredze, M., & Harman, C. (2014). Quantifying mental health signals in twitter. Proceedings of the Workshop on Computational Linguistics and Clinical Psychology: From Linguistic Signal to Clinical Reality, 51-60. https://doi.org/10.3115/v1/w14-3207
  14. De Choudhury, M. C. (2013). Social media as a measurement tool of depression in populations. Proceedings of the 5th Annual ACM Web Science Conference, 47-56. ACM. https://doi.org/10.1145/2464464.2464480
  15. De Choudhury, M., Gamon, M., Counts, S., & Horvitz, E. (2013). Predicting depression via social media. ICWSM, 13, 1-10.
  16. Hammen, C., & Brennan, P. A. (2002). Interpersonal dysfunction in depressed women: Impairments independent of depressive symptoms. Journal of Affective Disorders, 72(2), 145-156. https://doi.org/10.1016/s0165-0327(01)00455-4
  17. Jiang, L., & Yang, C. C. (2013). Using co-occurrence analysis to expand consumer health vocabularies from social media data. IEEE International Conference on Healthcare Informatics, 75-81. https://doi.org/10.1109/ichi.2013.16
  18. Mimno, D., & McCallum, A. (2008). Topic models conditioned on arbitrary features with dirichlet-multinomial regression. The 24th Conference on Unvertainty in Artificial Intelligence, 411-418. Helsinki: UAI.
  19. NAMI. (2014). Tell me about depression. Retrieved from NAMI - National Alliance on Mental Illness: https://www.nami.org/Learn-More/Mental-Health-Conditions/Depression
  20. Pang, B., Lee, L., & Vaithyanathan, S. (2002). Thumbs up?: sentiment classification using machine learning techniques. Proceedings of the ACL-02 conference on Empirical methods in natural language processing, 10, 79-86. Association for Computational Linguistics.
  21. Park, M., Cha, C., & Cha, M. (2012). Depressive moods of users portrayed in twitter. Proceedings of the ACM SIGKDD Workshop on Healthcare Informatics, 1-8. New York: ACM.
  22. SHINEWARE. (2013. 3. 20). Products-komoran. Retrieved from SHINEWARE: https://www.shineware.co.kr/products/komoran/
  23. Snow, R., O'Connor, B., Jurafsky, D., & Ng, Y. A. (2008). Cheap and fast---but is it good?: evaluating non-expert annotations for natural language tasks. Proceedings of the Conference on Empirical Methods in Natural Language Processing, 254-263. Association for Computational Linguistics. https://doi.org/10.3115/1613715.1613751
  24. Wang, H., Can, D., Kazemzadeh, A., Bar, F., & Narayanan, S. (2012). A system for real-time twitter sentiment analysis of 2012 us presidential election cycle. Proceedings of the ACL 2012 System Demonstrations, 115-120. Association for Computational Linguistics.
  25. Wang, Z. (2017). Machine learning methods for finding textual features of depression from publications. Georgia: Georgia State University.
  26. Zhao, W. X., Jiang, J., Weng, J., He, J., Lim, E. P., Yan, H., & Li, X. (2011). Comparing twitter and traditional media using topic models. European conference on information retrieval, 338-349. Berlin: Springer.
  27. Zhou, Z., Wang, W., & Wang, L. (2012). Community detection based on an improved modularity. Springer, 638-645. Berlin: Heidelberg. https://doi.org/10.1007/978-3-642-33506-8_78