{"id":"W4386869806","doi":"10.2139/ssrn.4567954","title":"Multi-View Representation Learning to Schema Match Databases","year":2023,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Centre for Addiction and Mental Health","funders":"","keywords":"Schema (genetic algorithms); Computer science; Database; Information retrieval; Representation (politics); Schema evolution; Database schema; Database design; Political science","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":[0.003406471,0.0009727033,0.002045598,0.003556213,0.0008698563,0.003644028,0.003590218,0.002408206,0.004884177],"category_scores_gemma":[0.01661277,0.0007603568,0.002334233,0.005786486,0.0008301684,0.007104672,0.003680507,0.002950178,0.002095949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001523085,"about_ca_system_score_gemma":0.001905613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00595158,"about_ca_topic_score_gemma":0.007131109,"domain_scores_codex":[0.9961745,0.001093864,0.0002984045,0.001199427,0.0009671255,0.0002666633],"domain_scores_gemma":[0.9922407,0.003380232,0.0003859398,0.002919191,0.0008089469,0.0002650136],"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":[0.0006480592,0.0007598121,0.008198251,0.0004319477,0.000493235,0.0001927489,0.00037925,0.05070714,0.005380304,0.02572522,0.02915732,0.8779267],"study_design_scores_gemma":[0.00004747242,0.0001594974,0.0008508659,0.00006207274,0.0001269426,0.0002331831,0.0002353318,0.917802,0.004672701,0.07086649,0.00492027,0.00002314718],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02873709,0.001522048,0.9614704,0.0009188764,0.0001426956,0.0001941634,0.001383316,0.004237636,0.001393758],"genre_scores_gemma":[0.3762591,0.0009325425,0.6106996,0.0007484809,0.0001638851,0.0002652415,0.007652444,0.0003766149,0.002902088],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00595158,"threshold_uncertainty_score":0.01801533,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07372381270653196,"score_gpt":0.3584526914609688,"score_spread":0.2847288787544368,"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."}}