{"id":"W4308835006","doi":"10.1002/pds.5555","title":"More extreme duplication in FDA Adverse Event Reporting System detected by literature reference normalization and fuzzy string matching","year":2022,"lang":"en","type":"article","venue":"Pharmacoepidemiology and Drug Safety","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Adverse Event Reporting System; Computer science; Data mining; Information retrieval; Levenshtein distance; Normalization (sociology); Medicine; Matching (statistics); String metric; String searching algorithm; Adverse effect; Artificial intelligence; Pattern matching; Internal medicine","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01954596,0.0006912838,0.001013935,0.01951627,0.00157053,0.00329939,0.002026611,0.001272497,0.002165277],"category_scores_gemma":[0.09270237,0.0004388493,0.00146017,0.01814211,0.0009010474,0.002736435,0.002362862,0.0009137431,0.001131564],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002077564,"about_ca_system_score_gemma":0.003826732,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007909824,"about_ca_topic_score_gemma":0.006116641,"domain_scores_codex":[0.9753092,0.00429558,0.005905763,0.004330624,0.009508662,0.0006501889],"domain_scores_gemma":[0.9050512,0.04132935,0.01922499,0.01239009,0.02120255,0.0008018304],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001129338,0.0004264558,0.3456759,0.002794162,0.0008773555,0.005701009,0.004801033,0.01757329,0.04284678,0.0113846,0.01566283,0.5511273],"study_design_scores_gemma":[0.0001904661,0.001118471,0.468954,0.001780514,0.002064439,0.01070465,0.00612439,0.1801038,0.1820654,0.02614254,0.1201819,0.0005694436],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7559217,0.00554836,0.2054762,0.002503789,0.0007045699,0.0009666324,0.01059322,0.005466421,0.01281901],"genre_scores_gemma":[0.7243852,0.001208581,0.2528357,0.0007239931,0.0001994149,0.0005044824,0.01675311,0.0003743068,0.003015347],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.980454,"threshold_uncertainty_score":0.1033701,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02566215284907491,"score_gpt":0.3131290756210826,"score_spread":0.2874669227720076,"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."}}