{"id":"W2715985536","doi":"10.2196/publichealth.6577","title":"Filtering Entities to Optimize Identification of Adverse Drug Reaction From Social Media: How Can the Number of Words Between Entities in the Messages Help?","year":2017,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Pharmacovigilance and Adverse Drug Reactions","field":"Pharmacology, Toxicology and Pharmaceutics","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Popularity; False positive paradox; Term (time); Social media; Recall; Drug; Computer science; Medicine; Information retrieval; Psychiatry; Psychology; Machine learning; Cognitive psychology; World Wide Web; Social psychology","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":[0.001964985,0.0001368757,0.0003202748,0.00007945234,0.0007032272,0.00005774354,0.0004190916,0.00009606644,0.00004867523],"category_scores_gemma":[0.0002761069,0.000102848,0.00006659107,0.000141027,0.0003016797,0.0003239542,0.00008186778,0.0004148898,0.000003971695],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005599758,"about_ca_system_score_gemma":0.000181877,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006159393,"about_ca_topic_score_gemma":0.0009312982,"domain_scores_codex":[0.9981838,0.000649476,0.0004030056,0.0002093924,0.0002193544,0.0003349868],"domain_scores_gemma":[0.9981798,0.0007929082,0.0005248637,0.0002736068,0.00009216847,0.0001366324],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001426332,0.0001758688,0.9174259,0.0002402259,0.0001441174,0.000003933024,0.05395755,0.00002027331,0.003284209,0.0006584427,0.008863444,0.01508339],"study_design_scores_gemma":[0.0009773887,0.0000122559,0.9197038,0.00002007304,0.000009654167,0.000002582792,0.01836153,0.00008489656,0.0006725576,0.0002253721,0.05977318,0.0001567203],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9626952,0.0001373878,0.00002344026,0.03486005,0.0006434145,0.0004004369,0.0005634835,0.00001831419,0.000658221],"genre_scores_gemma":[0.9974965,0.0008708535,0.00001461136,0.0007131714,0.0003816474,0.0001284562,0.00009937837,0.000009343677,0.000286074],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05090974,"threshold_uncertainty_score":0.5408726,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1290499107859054,"score_gpt":0.437439936973234,"score_spread":0.3083900261873286,"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."}}