{"id":"W3201584721","doi":"10.1016/j.dib.2021.107360","title":"Dataset of COVID-19 outbreak and potential predictive features in the USA","year":2021,"lang":"en","type":"article","venue":"Data in Brief","topic":"COVID-19 epidemiological studies","field":"Mathematics","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Brock University","funders":"Alberta Innovates; Isfahan University of Technology; Pfizer","keywords":"Outbreak; Coronavirus disease 2019 (COVID-19); Pandemic; Social distance; Public health; 2019-20 coronavirus outbreak; Geography; Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); Disease; Data science; Medicine; Computer science; Virology; Infectious disease (medical specialty)","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007887047,0.0004935996,0.0004302883,0.001786232,0.0004955468,0.0007326961,0.0009234911,0.001083198,0.008078046],"category_scores_gemma":[0.005829342,0.0002075027,0.0005052172,0.003397901,0.0002076327,0.0005314,0.0007940128,0.001207338,0.003853888],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007977164,"about_ca_system_score_gemma":0.001441821,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04704271,"about_ca_topic_score_gemma":0.07419058,"domain_scores_codex":[0.9995483,0.0001086091,0.0000575964,0.0001086873,0.0001183791,0.00005854471],"domain_scores_gemma":[0.997582,0.0009934088,0.0002471015,0.0002912156,0.0007024151,0.0001838493],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.0001969932,0.0003035616,0.04788207,0.0002546342,0.00008711836,0.0002015704,0.0001040504,0.005875123,0.0003152002,0.00176713,0.925387,0.01762555],"study_design_scores_gemma":[0.0004856769,0.0002111685,0.235839,0.000553137,0.0001125674,0.0005256655,0.001108987,0.05199771,0.001040271,0.007814912,0.7001944,0.000116582],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03373379,0.0003354301,0.001107243,0.001294996,0.0001585538,0.00008164331,0.9592415,0.0004706253,0.003576235],"genre_scores_gemma":[0.03550365,0.0001745794,0.001658266,0.0002150021,0.00004545988,0.0001547693,0.9611623,0.00002663012,0.001059374],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.04704271,"threshold_uncertainty_score":0.09353775,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2212606958055259,"score_gpt":0.440343947454661,"score_spread":0.2190832516491351,"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."}}