{"id":"W3047131763","doi":"10.2196/20794","title":"Big Data, Natural Language Processing, and Deep Learning to Detect and Characterize Illicit COVID-19 Product Sales: Infoveillance Study on Twitter and Instagram","year":2020,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Pharmaceutical Quality and Counterfeiting","field":"Medicine","cited_by":120,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Social media; Counterfeit; Internet privacy; Public health; Sentiment analysis; Pandemic; Advertising; The Internet; Computer science; Public health surveillance; Product (mathematics); World Wide Web; Coronavirus disease 2019 (COVID-19); Business; Medicine; Artificial intelligence; Political science; 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":[],"consensus_categories":[],"category_scores_codex":[0.001339112,0.0002382856,0.0005202899,0.00009170215,0.00029977,0.0002132898,0.0001126102,0.00005441321,0.000003975975],"category_scores_gemma":[0.00210752,0.0001987056,0.00001261745,0.0003105257,0.0001201737,0.0001907717,0.0002353956,0.0005648417,0.000002822117],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004259105,"about_ca_system_score_gemma":0.0002764759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006312754,"about_ca_topic_score_gemma":0.0001086583,"domain_scores_codex":[0.9976815,0.0003298645,0.0003895986,0.0007844326,0.000295335,0.0005192743],"domain_scores_gemma":[0.9977844,0.0002110798,0.000142734,0.0002496852,0.0000635827,0.001548484],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.000564086,0.00006694813,0.3515803,0.00218032,0.00002951566,0.00003906518,0.02984482,1.611834e-7,0.0004615026,0.000004421832,0.0001722228,0.6150566],"study_design_scores_gemma":[0.003902433,0.002107776,0.7445152,0.00009718886,0.000004745371,0.0001294519,0.01248926,0.002447653,0.000005898004,0.000003339214,0.2338216,0.000475416],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9424086,0.00620799,0.0001207182,0.04986125,0.00005062807,0.001138336,0.00002866192,0.0001307621,0.00005311769],"genre_scores_gemma":[0.9346429,0.0004887464,0.00007492748,0.06426763,0.0003344376,0.00004745223,0.00008308659,0.00002117467,0.00003970627],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6145812,"threshold_uncertainty_score":0.8102981,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1509587901491284,"score_gpt":0.4118671109403815,"score_spread":0.260908320791253,"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."}}