{"id":"W2949244208","doi":"10.48550/arxiv.1702.00820","title":"HoloClean: Holistic Data Repairs with Probabilistic Inference","year":2017,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":57,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Inference; Leverage (statistics); Probabilistic logic; Tuple; Computer science; Data mining; Statistical model; Machine learning; Artificial intelligence; Mathematics","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":["metaresearch","metaepi_narrow","open_science","insufficient_payload"],"consensus_categories":["open_science"],"category_scores_codex":[0.003680598,0.0004194346,0.0006528821,0.000358017,0.0004591253,0.00102668,0.01166686,0.0002443909,0.000281077],"category_scores_gemma":[0.01059419,0.0003541278,0.0001242865,0.000468169,0.0008459048,0.001139695,0.01474155,0.0006920684,0.0008949742],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001374936,"about_ca_system_score_gemma":0.0004647076,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007380922,"about_ca_topic_score_gemma":0.001444976,"domain_scores_codex":[0.9950578,0.0004415312,0.0004925678,0.002946459,0.0006071259,0.000454574],"domain_scores_gemma":[0.9839115,0.001026212,0.001020568,0.0131071,0.0006644822,0.0002701536],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000753959,0.0007956108,0.0211414,0.0005729541,0.0006210731,0.003343929,0.0007362921,0.2402453,0.000001256044,0.6308419,0.08788745,0.01305893],"study_design_scores_gemma":[0.001268173,0.0002420746,0.01810486,0.0005265116,0.0007237167,0.000007181783,0.001388439,0.229353,0.00000298031,0.6227182,0.1239994,0.001665439],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3308181,0.0001292575,0.5481998,0.001217144,0.003788243,0.002882492,0.004209274,0.000843262,0.1079124],"genre_scores_gemma":[0.9861795,0.00009982791,0.0006055864,0.0001266391,0.0001215041,0.000001593792,0.0003964268,0.00002156323,0.01244731],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6553614,"threshold_uncertainty_score":0.999891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.631797409494457,"score_gpt":0.3557862297306885,"score_spread":0.2760111797637684,"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."}}