{"id":"W1998597078","doi":"10.1145/1670243.1670248","title":"An information-theoretic analysis of worst-case redundancy in database design","year":2008,"lang":"en","type":"article","venue":"ACM Transactions on Database Systems","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Seventh Framework Programme; Engineering and Physical Sciences Research Council","keywords":"Computer science; Redundancy (engineering); Schema (genetic algorithms); Functional dependency; Database schema; Theoretical computer science; Data integrity; Relational database; Data mining; Database design; Information retrieval; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009919658,0.0003148786,0.0006242088,0.001704416,0.0003461944,0.0000622431,0.0009734325,0.00008703383,0.00004458877],"category_scores_gemma":[0.0001311117,0.0002967949,0.0001462591,0.003512766,0.0001699501,0.005274704,0.00004229934,0.0002793641,0.00004675517],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001071764,"about_ca_system_score_gemma":0.0001853559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00230234,"about_ca_topic_score_gemma":0.0003577521,"domain_scores_codex":[0.9968585,0.0005061816,0.001117669,0.0005644325,0.0005468594,0.0004063369],"domain_scores_gemma":[0.9946991,0.0004707822,0.0003570997,0.004064915,0.000176656,0.0002314862],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004529611,0.002147634,0.0009720215,0.0009529034,0.001479815,0.004161713,0.01190788,0.8090082,0.002775057,0.1492956,0.0006947381,0.01615145],"study_design_scores_gemma":[0.002771359,0.0006807006,0.0005085507,0.0008077925,0.000634247,0.003970108,0.003683454,0.9704953,0.006467935,0.0001429388,0.008164986,0.001672591],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01178142,0.0001537483,0.9843577,0.00004038466,0.0004133509,0.0006364483,0.002379375,0.0001608745,0.0000766585],"genre_scores_gemma":[0.8550411,0.0001582552,0.1438545,0.00005012932,0.00002294237,0.0001772202,0.0006477304,0.0000162984,0.00003182571],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8432597,"threshold_uncertainty_score":0.9999484,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03518479320693964,"score_gpt":0.275027395432391,"score_spread":0.2398426022254513,"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."}}