Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Capturing social media signals for health research

246 views

Published on

Presented at Digital Demography workshop at Max-Planck Institute for Demographic Research on Oct 17, 2018

Published in: Data & Analytics
  • Thankfully this community blesses you with the right information and tools to get through the inevitable tough times and the joy of having people who are in the same situation and helping you as you go. I'm forever thankful for having the irreplaceable support to fight for what is best for me with all force and knowledge and much love! Shaye and other girls did it, I'm doing it and do can anyone struggling from this, no matter how deep and dark it seems! One step at a time, TOGETHER!  http://t.cn/A6Pq6OB6
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
  • Be the first to like this

Capturing social media signals for health research

  1. 1. capturing social media signals for health research Yelena Mejova Previously:
  2. 2. 2 Time & date Location Photo Explanation Social response Ramadan Food Exercise Sentiment Wearable Phone
  3. 3. 3 Annie S. Anderson, (1995) British Food Journal, Vol. 97 Issue: 7, pp.22-26
  4. 4. 4 1995 Newsgroups 2000 Websites 2005 Social Media 2008 Smartphones 2013 Wearables 2017 Drones, VR, Genomics…
  5. 5. self-motivated plentiful real-time geo-located media rich social cultural interactive 5 self-image noisy bursty geo-biased complex signal influence contextual persuasive
  6. 6. Text Images Location Social Wearables Search Advertising 6
  7. 7. Text (Blogs, Microblogs, Comments, News…) • Track influenza-like-illness (ILI) symptoms • Global Epidemic and Mobility Model (GLEAM) • Predict Influenza seasons 7Zhang+ WWW’17 *ISI GPS Located
  8. 8. Text (Blogs, Microblogs, Comments, News…) • Food-related queries to Instagram • Relating foods in tags to “food deserts” 8De Choudhury+ CSCW’16 nutritional attributes of food deserts and non-food deserts
  9. 9. Text (Blogs, Microblogs, Comments, News…) • receiving social support in the first post leads to a relative increase in the achieved weight loss of 26%, or an absolute mean difference of 9 lbs. 9 loss age/gender height starting weight current weight goal weight Cunha+ WWW’17 *QCRI
  10. 10. Images (Flickr, Instagram…) 10 https://www.kairos.com/demos https://imagga.com/ Garimella+ CHI’16 *QCRI • using labels extracted from images allows for better tracking of certain conditionsPearson’s r for predicting health statistics across 100 counties (U)ser tags, (I)magga tags, (D)emographics
  11. 11. Images (Flickr, Instagram…) 11Kocabey+ ICWSM’17 *QCRI BMI (Body Mass Index) • using deep learning on visual features works nearly as well as human labeling
  12. 12. Location (metadata) 12Mejova+ DH’15 • are check-ins at fast food restaurants associated with higher risk of obesity? • is there different behavior between low and high obesity areas?
  13. 13. • hybrid model of human mobility integrating Flickr data with the classical gravity model 13Beiró+ EPJ Data Science’16 *ISI Location (metadata)
  14. 14. Social (friendships, likes, comments…) 14Mejova+ ICWSM’17 • social perception of #foodporn, response to different framing #foodporn
  15. 15. Social (friendships, likes, comments…) 15Aral+Nicolaides Nature Communications’17 • social network for runners • measuring influence of runners on their friends’ running performance • using weather as instrumental variable to “randomize” data
  16. 16. Wearables (shared on SM) 16Wang+ DH’16, Akbar+ ICHI’16 *QCRI • building models to predict weight • comparing sleeping patterns across the world
  17. 17. Wearables (+ Games!) 17Althoff+ JMIR’16 • detect Pokemon Go users through “experiential queries” • link to Microsoft Band pedometer info • users issuing at least 10 queries (very interested in game) added 1473 daily steps in first week
  18. 18. Search Queries (ok, not quite SM) 18 Google Flu Trends Ginsberg+ Nature’09, Nuti+, Ojala+ ICWSM’17, PLOS’14, Yom-Tov+ JMIR’14 Bing for Tracking Mood Disorders Google Trends in 2014: infectious disease (27% of articles), mental health and substance use (24%), other non-communicable diseases (16%), and general population behavior (33%). By use, 27% of articles utilized Google Trends for casual inference, 39% for description, and 34% for surveillance. Google Correlate for Fertility
  19. 19. Advertisement (on SM platforms) see talks by Ingmar Weber & Ridhi Kashyap 19
  20. 20. 360o Epidemiology Tracking disease and its awareness in demographic, cultural, social, … context 20 Modeling health behavior Social influence Media influence Technology influence Campaigns and interventions Connecting with health orgs Building tools with population and individual data Data-driven intervention
  21. 21. • Zhang, Q., Perra, N., Perrotta, D., Tizzoni, M., Paolotti, D., & Vespignani, A. (2017). Forecasting seasonal influenza fusing digital indicators and a mechanistic disease model. International Conference on World Wide Web. web • De Choudhury, Munmun, Sanket Sharma, and Emre Kiciman. "Characterizing dietary choices, nutrition, and language in food deserts via social media." Proceedings of the 19th acm conference on computer-supported cooperative work & social computing. ACM, 2016. web • Cunha, Tiago, Ingmar Weber, and Gisele Pappa. "A warm welcome matters!: The link between social feedback and weight loss in/r/loseit." Proceedings of the 26th International Conference on World Wide Web Companion. International World Wide Web Conferences Steering Committee, 2017. web • Garimella, Venkata Rama Kiran, Abdulrahman Alfayad, and Ingmar Weber. "Social media image analysis for public health." Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems. ACM, 2016. web • Kocabey, Enes, et al. "Face-to-BMI: using computer vision to infer body mass index on social media." ICWSM (2017). web • Mejova, Y., Haddadi, H., Noulas, A., & Weber, I. (2015, May). # foodporn: Obesity patterns in culinary interactions. In Proceedings of the 5th International Conference on Digital Health 2015 (pp. 51-58). ACM. web • Beiró, M. G., Panisson, A., Tizzoni, M., & Cattuto, C. (2016). Predicting human mobility through the assimilation of social media traces into mobility models. EPJ Data Science, 5(1), 30. web • Mejova, Y., Abbar, S., & Haddadi, H. (2016, May). Fetishizing Food in Digital Age:# foodporn Around the World. In ICWSM. web • Aral, Sinan, and Christos Nicolaides. "Exercise contagion in a global social network." Nature communications 8 (2017): 14753. web • Wang, Y., Weber, I., & Mitra, P. (2016, April). Quantified self meets social media: sharing of weight updates on Twitter. In Proceedings of the 6th International Conference on Digital Health Conference (pp. 93-97). ACM. web • Akbar, F., & Weber, I. (2016, October). # Sleep_as_Android: Feasibility of Using Sleep Logs on Twitter for Sleep Studies. In Healthcare Informatics (ICHI), 2016 IEEE International Conference on (pp. 227-233). IEEE. web • Althoff, T., White, R. W., & Horvitz, E. (2016). Influence of Pokémon Go on physical activity: study and implications. Journal of medical Internet research, 18(12). web • Ginsberg, J., Mohebbi, M. H., Patel, R. S., Brammer, L., Smolinski, M. S., & Brilliant, L. (2009). Detecting influenza epidemics using search engine query data. Nature, 457(7232), 1012. web • Yom-Tov, E., White, R. W., & Horvitz, E. (2014). Seeking insights about cycling mood disorders via anonymized search logs. Journal of medical Internet research, 16(2). web • Nuti, S. V., Wayda, B., Ranasinghe, I., Wang, S., Dreyer, R. P., Chen, S. I., & Murugiah, K. (2014). The use of google trends in health care research: a systematic review. PloS one, 9(10), e109583. web • Ojala, J., Zagheni, E., Billari, F. C., & Weber, I. (2017). Fertility and its meaning: Evidence from search behavior. ICWSM. web • Abbar, S., Mejova, Y., & Weber, I. (2015, April). You tweet what you eat: Studying food consumption through twitter. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (pp. 3197-3206). ACM. web • Christakis, N. A., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years. New England journal of medicine, 357(4), 370-379. web 21
  22. 22. 22 @yelenamejova yelenamejova.com * Previously

×