{"id":"W2157631542","doi":"10.1145/1807167.1807213","title":"Schema clustering and retrieval for multi-domain pay-as-you-go data integration systems","year":2010,"lang":"en","type":"article","venue":"","topic":"Semantic Web and Ontologies","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Data integration; Cluster analysis; Schema (genetic algorithms); Data mining; Information retrieval; Conceptual schema; Schema matching; Probabilistic logic; Machine learning; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008950927,0.0007184206,0.001472792,0.005275405,0.002173917,0.00370252,0.00325493,0.002120103,0.002274744],"category_scores_gemma":[0.01891039,0.0008055627,0.001516486,0.005814392,0.001082296,0.00766298,0.003559076,0.001684525,0.001386939],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003040511,"about_ca_system_score_gemma":0.002598654,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007273044,"about_ca_topic_score_gemma":0.009239788,"domain_scores_codex":[0.9923153,0.002258501,0.0008557061,0.001303338,0.002947408,0.0003197543],"domain_scores_gemma":[0.9897193,0.002834649,0.0008388329,0.004092633,0.002242326,0.0002723206],"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":[0.00061063,0.0007059895,0.009400334,0.0006375117,0.000382828,0.0003841086,0.002102494,0.09479935,0.02433831,0.0718,0.0159954,0.778843],"study_design_scores_gemma":[0.00009194559,0.000127259,0.002708317,0.00007148666,0.0001117047,0.0006593585,0.001033316,0.9052064,0.0251089,0.04261769,0.02217317,0.00009040606],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01891065,0.0004455352,0.9724691,0.0003965829,0.00002335951,0.0004069275,0.0002636742,0.005738872,0.001345446],"genre_scores_gemma":[0.1429645,0.000287704,0.8532584,0.0002233884,0.00002407024,0.0002876616,0.001544255,0.0002778234,0.001132227],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008950927,"threshold_uncertainty_score":0.04733765,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07672508189442405,"score_gpt":0.3239843287210066,"score_spread":0.2472592468265825,"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."}}