{"id":"W4393327679","doi":"10.1016/j.knosys.2024.111678","title":"Better entity matching with transformers through ensembles","year":2024,"lang":"en","type":"article","venue":"Knowledge-Based Systems","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"National Research Council Canada; McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Transformer; Computer science; Matching (statistics); Artificial intelligence; Data mining; Engineering; Mathematics; Electrical engineering; Statistics; Voltage","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003872568,0.0002512492,0.0004017767,0.0003029575,0.0002353578,0.001615851,0.0006982654,0.00008521773,0.0002661247],"category_scores_gemma":[0.00008541359,0.0001603376,0.0001752435,0.001002575,0.0001279573,0.0009036984,0.00006392358,0.0002009285,0.003230161],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009789805,"about_ca_system_score_gemma":0.0001736116,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000407899,"about_ca_topic_score_gemma":0.0008232193,"domain_scores_codex":[0.9964525,0.0004691978,0.0007287091,0.0007605383,0.001191566,0.000397461],"domain_scores_gemma":[0.9979576,0.001001006,0.00009466302,0.0006765994,0.0001597964,0.0001102881],"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":[0.0003723997,0.0008192873,0.001536947,0.005229124,0.0008254455,0.0004929538,0.02819705,0.002912121,0.00225592,0.2264928,0.5488985,0.1819675],"study_design_scores_gemma":[0.0004752504,0.0001070839,0.0001449383,0.0006475974,0.00006737745,0.00001008949,0.00391599,0.004971927,0.0008763977,0.004840201,0.9835816,0.0003615619],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06334736,0.003784275,0.8274451,0.002554261,0.004229384,0.0008580278,0.0001852625,0.0004497901,0.09714649],"genre_scores_gemma":[0.9917038,0.00001007286,0.0005220514,0.0002853104,0.0002870918,0.00007446808,0.0000347425,0.00002946965,0.007053004],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9283564,"threshold_uncertainty_score":0.9994206,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.122142341677316,"score_gpt":0.3903886435702951,"score_spread":0.2682463018929792,"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."}}