{"id":"W4252876123","doi":"10.2196/preprints.32355","title":"Physical Activity, Sedentary Behavior, and Sleep on Twitter: Multicountry and Fully Labeled Public Data Set for Digital Public Health Surveillance Research (Preprint)","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Social media; Computer science; Metadata; Data set; Public health; Public health surveillance; Set (abstract data type); Data quality; Data science; World Wide Web; Medicine; Engineering; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.003062067,0.0006688644,0.00151765,0.000410327,0.0003026565,0.001684111,0.001025429,0.000348334,0.00009689763],"category_scores_gemma":[0.0021074,0.0006314215,0.0001583877,0.0003765836,0.0005361327,0.0008343034,0.009196598,0.001969978,0.00002208035],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004478424,"about_ca_system_score_gemma":0.001772833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007295919,"about_ca_topic_score_gemma":0.001043933,"domain_scores_codex":[0.9927659,0.0006739429,0.0005665108,0.003066993,0.001618363,0.001308273],"domain_scores_gemma":[0.9919628,0.001238027,0.0002974581,0.004329229,0.0009074382,0.001265049],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001336748,0.008257962,0.6680917,0.00594217,0.001369522,0.0003885367,0.0008810887,0.000004361708,0.0005419367,0.0001815131,0.0353613,0.2776432],"study_design_scores_gemma":[0.007984976,0.001035635,0.9491152,0.0006573368,0.000126701,0.0001870575,0.001136128,0.0164496,0.0001037115,0.0001275194,0.02166351,0.001412623],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9633563,0.0006320969,0.0004036954,0.01544516,0.00029802,0.004749056,0.01454126,0.0002921966,0.0002821694],"genre_scores_gemma":[0.9678673,0.0004914633,0.0006784779,0.0007910109,0.0004522677,0.0005823208,0.02875692,0.0001256905,0.0002545324],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2810236,"threshold_uncertainty_score":0.9996137,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.257005036698103,"score_gpt":0.443589629380032,"score_spread":0.186584592681929,"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."}}