{"id":"W4399174990","doi":"10.1145/3654984","title":"Unstructured Data Fusion for Schema and Data Extraction","year":2024,"lang":"en","type":"article","venue":"Proceedings of the ACM on Management of Data","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Information retrieval; Schema (genetic algorithms); Component (thermodynamics); Data mining; Table (database)","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":[0.004862013,0.001593387,0.001557241,0.006114514,0.001077072,0.002766668,0.002541584,0.001382983,0.00415175],"category_scores_gemma":[0.01821226,0.0008305815,0.00274288,0.00765215,0.001160615,0.007479253,0.004695519,0.002757436,0.003603161],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001554175,"about_ca_system_score_gemma":0.002681629,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00352582,"about_ca_topic_score_gemma":0.005729317,"domain_scores_codex":[0.9950709,0.001003211,0.0005477504,0.00112372,0.002064773,0.0001896531],"domain_scores_gemma":[0.9894502,0.003347657,0.0006868435,0.004246322,0.002065433,0.0002035431],"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":[0.0004036581,0.0002785662,0.005788533,0.001537492,0.0003448768,0.0008471529,0.00110832,0.04544663,0.02907532,0.05368097,0.0351611,0.8263273],"study_design_scores_gemma":[0.00008141572,0.0002292187,0.00245652,0.000295901,0.0001905113,0.001447118,0.0009068243,0.6274142,0.08410067,0.1404037,0.1423143,0.0001596038],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005283117,0.001013606,0.9790235,0.0004291187,0.00009096947,0.0002791884,0.003824016,0.008567123,0.001489474],"genre_scores_gemma":[0.04773195,0.0006501893,0.9341906,0.0002889132,0.00004926557,0.0002496573,0.01518842,0.0004975746,0.001153388],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006114514,"threshold_uncertainty_score":0.02571309,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1355387895140462,"score_gpt":0.3611646660040275,"score_spread":0.2256258764899813,"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."}}