{"id":"W3032372024","doi":"10.2196/19509","title":"Machine Learning to Detect Self-Reporting of Symptoms, Testing Access, and Recovery Associated With COVID-19 on Twitter: Retrospective Big Data Infoveillance Study","year":2020,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Mental Health via Writing","field":"Psychology","cited_by":155,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Pandemic; Coronavirus disease 2019 (COVID-19); Computer science; Topic model; Public health; Artificial intelligence; Social media; Public health surveillance; Medicine; Machine learning; Natural language processing; Information retrieval; World Wide Web; Disease; Infectious disease (medical specialty)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.005806216,0.0003224144,0.0009506556,0.0001830445,0.0005038298,0.0001480056,0.0004230818,0.0001081602,0.0000192667],"category_scores_gemma":[0.0184651,0.0002948279,0.00002053306,0.00132111,0.00006293521,0.0002182272,0.0004921475,0.000699323,0.000003888609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003281654,"about_ca_system_score_gemma":0.0008505032,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001452839,"about_ca_topic_score_gemma":0.001067826,"domain_scores_codex":[0.994647,0.001162144,0.001611796,0.00118379,0.0004950862,0.0009002358],"domain_scores_gemma":[0.993737,0.001553038,0.002461246,0.000584841,0.000188468,0.001475449],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0002714072,0.000131221,0.955238,0.0003869325,0.00008735852,0.00003768943,0.006082606,0.00001036071,0.00000388651,0.00001045109,0.0004846302,0.0372555],"study_design_scores_gemma":[0.0022105,0.005248169,0.9845054,0.00008363647,0.000002349499,0.00003888935,0.002182673,0.000913742,3.397369e-7,0.00001165636,0.004473327,0.0003293508],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9876743,0.000487628,0.0002273985,0.00757216,0.0001647242,0.001990527,0.0001370417,0.0003401453,0.00140605],"genre_scores_gemma":[0.9878413,0.00004325484,0.0002836865,0.01134285,0.0001826097,0.0001176066,0.0001068392,0.00005116878,0.00003066199],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03692615,"threshold_uncertainty_score":0.9999504,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1749465264901289,"score_gpt":0.4103833435911393,"score_spread":0.2354368171010104,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}