{"id":"W3156937620","doi":"10.1101/2021.04.13.21255449","title":"Physical activity, sedentary behaviour, and sleep on Twitter: A labelled dataset for public health research","year":2021,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Agency of Canada; University of Calgary","funders":"","keywords":"Demographics; Social media; Public health; Public health surveillance; Computer science; Physical activity; Data science; Medicine; Internet privacy; Artificial intelligence; World Wide Web; Demography; Sociology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.001057494,0.0007246013,0.0006008745,0.002494445,0.0008267538,0.0007976445,0.0009445008,0.001875089,0.004439456],"category_scores_gemma":[0.005865925,0.0002306373,0.0006742332,0.002452141,0.0005995787,0.0007191048,0.001571942,0.00088645,0.00426375],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001225276,"about_ca_system_score_gemma":0.001145271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03187772,"about_ca_topic_score_gemma":0.06051252,"domain_scores_codex":[0.9989194,0.0003556364,0.0001049946,0.0002675766,0.0001977632,0.0001545891],"domain_scores_gemma":[0.9968893,0.001329246,0.0004434086,0.00056628,0.00044396,0.0003278038],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001774295,0.0006885775,0.1862831,0.004112978,0.0004469379,0.00194803,0.001580029,0.009988576,0.009262077,0.003020915,0.681683,0.09921145],"study_design_scores_gemma":[0.0004649301,0.0005048536,0.5263561,0.001165848,0.0002964872,0.001644287,0.003672392,0.04478893,0.006542531,0.005551356,0.408753,0.0002592827],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1142033,0.001184215,0.002821821,0.002565189,0.0003724825,0.0003519997,0.8728857,0.0009122537,0.004703098],"genre_scores_gemma":[0.1151301,0.0004343909,0.006568878,0.0004965661,0.000275702,0.0006030986,0.8728712,0.00007159378,0.003548421],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03187772,"threshold_uncertainty_score":0.06338435,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2459479852636265,"score_gpt":0.4482997752509828,"score_spread":0.2023517899873563,"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."}}