{"id":"W1971778458","doi":"10.1145/773153.773155","title":"An information-theoretic approach to normal forms for relational and XML data","year":2003,"lang":"en","type":"article","venue":"","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Relational database; XML; Normalization (sociology); Relational model; Information retrieval; Database normalization; Data mining; Measure (data warehouse); XML validation; Set (abstract data type); Theoretical computer science; Programming language; Artificial intelligence; Cluster analysis; World Wide Web","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003451912,0.00006641904,0.00006791712,0.00004767811,0.0001296238,0.00007025716,0.0002741252,0.00002193462,0.000005381761],"category_scores_gemma":[0.0000835446,0.00004980675,0.000007951569,0.0001078816,0.00002201592,0.005589715,0.0001318304,0.00002842764,0.0000111446],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000006906362,"about_ca_system_score_gemma":0.00004124606,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005337008,"about_ca_topic_score_gemma":0.000004007939,"domain_scores_codex":[0.9993955,0.00001280653,0.0001626913,0.0001805166,0.0001140602,0.0001343748],"domain_scores_gemma":[0.9991511,0.00004169946,0.00003608687,0.000631534,0.00005066453,0.00008888017],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000002497114,0.000009177455,0.00008301294,0.00001318801,0.000001950312,3.650792e-8,0.0002351458,0.000193243,0.000004975548,0.9957833,0.0005750977,0.003098317],"study_design_scores_gemma":[0.000554969,0.0001306665,0.0006680902,0.000008533213,0.000003729702,0.00005281315,0.0004710894,0.3018848,0.0002112372,0.01749555,0.6782143,0.0003041971],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000635458,0.00001009079,0.981013,0.00007825252,0.0000580818,0.0002604656,0.00008709004,0.00004877693,0.0178088],"genre_scores_gemma":[0.07112478,0.000001274121,0.9280429,0.0003590478,0.00001631994,0.00003322753,0.0002998351,0.000002783773,0.00011983],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9782878,"threshold_uncertainty_score":0.4052408,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02823130801210522,"score_gpt":0.2619315374141282,"score_spread":0.233700229402023,"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."}}