{"id":"W2405050116","doi":"10.1016/j.jval.2016.03.1777","title":"PROBABILISTIC RECORD MATCHING USING MACHINE LEARNING TECHNIQUES","year":2016,"lang":"en","type":"article","venue":"Value in Health","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Western University","funders":"","keywords":"Computer science; Matching (statistics); Data mining; Weighting; Confusion matrix; Set (abstract data type); Metric (unit); Probabilistic logic; Levenshtein distance; Artificial intelligence; Algorithm; Mathematics; Statistics","routes":{"ca_aff":true,"ca_fund":false,"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":[0.01231366,0.0005183887,0.001922687,0.007096412,0.0015418,0.003819946,0.003810086,0.002559528,0.006954714],"category_scores_gemma":[0.07949434,0.0009177138,0.00271677,0.01115317,0.0008383995,0.007273553,0.003354746,0.002117704,0.001936093],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001337566,"about_ca_system_score_gemma":0.003681749,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003857303,"about_ca_topic_score_gemma":0.004099174,"domain_scores_codex":[0.9808967,0.006946649,0.0025806,0.003097453,0.005873047,0.0006056132],"domain_scores_gemma":[0.9494806,0.0355022,0.003280869,0.006964979,0.004363268,0.0004081657],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000639306,0.000496543,0.01766399,0.0007101074,0.000447204,0.000329233,0.0004026992,0.08708625,0.003343693,0.05449773,0.008507355,0.8258759],"study_design_scores_gemma":[0.00007371393,0.0001493021,0.003102974,0.0001070455,0.0002419521,0.0005864848,0.0001470308,0.8598331,0.006678423,0.1203806,0.008633477,0.00006595687],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01240158,0.0004160622,0.9837754,0.0004435701,0.00008036211,0.0002201833,0.0006424412,0.001024624,0.000995733],"genre_scores_gemma":[0.2643044,0.00058151,0.7283452,0.0001960792,0.0001605502,0.0003961158,0.002534373,0.0001943187,0.003287491],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01231366,"threshold_uncertainty_score":0.06512165,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3116090154795184,"score_gpt":0.451216637542168,"score_spread":0.1396076220626496,"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."}}