Here are the slides for my tutorial on applying causal inference to social media timelines, presented at the Workshop on Observational Studies in Social Media (OSSM) at ICWSM 2017. The talk emphasizes counterfactual intuitions and highlights common pitfalls. This tutorial steps through the process of analyzing social media timelines with a focus on treatment identification, covariate featurization and outcome analysis applicable to a broad class of conditioned inference algorithms. tutorial_kiciman_ossm17
Here are the papers I referenced in the slides:
- Towards Decision Support and Goal Achievement: Identifying Action-Outcome Relationships from Social Media. Kıcıman, Richardson. KDD15
- Shifts to Suicidal Ideation from Mental Health Content in Social Media. De Choudhury, Kıcıman, Dredze, Coppersmith, Kumar. CHI16
- Distilling the Outcomes of Personal Experiences: A Propensity-scored Analysis of Social Media. Olteanu, Varol, Kıcıman. CSCW17
- The Language of Social Support in Social Media and its Effect on Suicidal Ideation Risk, De Choudhury, Kıcıman ICWSM17
- Using Social Media to Understand the Effects of Alcohol Use During College. Kıcıman, Counts, Gasser (email for draft)
- Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries. Olteanu, Castillo, Diaz and Kıcıman Working paper