AMIA Jt Summits Transl Sci Proc. 2015 Mar 25;2015:122-6. eCollection 2015.

Mining Twitter Data to Improve Detection of Schizophrenia.

AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science

Kimberly McManus, Emily K Mallory, Rachel L Goldfeder, Winston A Haynes, Jonathan D Tatum

PMID: 26306253 PMCID: PMC4525233

Abstract

Individuals who suffer from schizophrenia comprise I percent of the United States population and are four times more likely to die of suicide than the general US population. Identification of at-risk individuals with schizophrenia is challenging when they do not seek treatment. Microblogging platforms allow users to share their thoughts and emotions with the world in short snippets of text. In this work, we leveraged the large corpus of Twitter posts and machine-learning methodologies to detect individuals with schizophrenia. Using features from tweets such as emoticon use, posting time of day, and dictionary terms, we trained, built, and validated several machine learning models. Our support vector machine model achieved the best performance with 92% precision and 71% recall on the held-out test set. Additionally, we built a web application that dynamically displays summary statistics between cohorts. This enables outreach to undiagnosed individuals, improved physician diagnoses, and destigmatization of schizophrenia.

References

1. Schizophr Bull. 2008 Sep;34(5):819-34 - PubMed
2. Am J Psychiatry. 1982 Mar;139(3):275-82 - PubMed
3. J Psychopharmacol. 2010 Nov;24(4 Suppl):81-90 - PubMed
4. Curr Dir Psychol Sci. 2010 Aug;19(4):255-259 - PubMed
5. Health Serv Res. 2001 Dec;36(6 Pt 1):987-1007 - PubMed

Publication Types

• Journal Article

Grant support

• T15 LM007033/LM/NLM NIH HHS/United States
• T32 GM007276/GM/NIGMS NIH HHS/United States