{"id":"W4402806881","doi":"10.1145/3696110","title":"Automated anomaly detection for categorical data by repurposing a form filling recommender system","year":2024,"lang":"en","type":"article","venue":"Journal of Data and Information Quality","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Fonds National de la Recherche Luxembourg; Natural Sciences and Engineering Research Council of Canada; BNP Paribas Cardif; Canada Research Chairs; Science Foundation Ireland","keywords":"Repurposing; Computer science; Categorical variable; Recommender system; Anomaly detection; Data mining; Anomaly (physics); Artificial intelligence; Information retrieval; Machine learning","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":[0.00242391,0.001020686,0.001538939,0.003489438,0.0006471191,0.00130454,0.002395192,0.001345336,0.001276666],"category_scores_gemma":[0.01381316,0.0004280077,0.001232573,0.002430668,0.0004614303,0.002077238,0.001326919,0.001516042,0.00154038],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007778803,"about_ca_system_score_gemma":0.001373397,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01157072,"about_ca_topic_score_gemma":0.01557805,"domain_scores_codex":[0.9973292,0.0004796236,0.0002757277,0.0007873395,0.0009555171,0.0001725847],"domain_scores_gemma":[0.9921741,0.002835057,0.0008200105,0.001679924,0.00221385,0.0002770142],"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":[0.000335739,0.0003959139,0.07709959,0.0002385712,0.0001699851,0.000398272,0.0004141322,0.02856445,0.01842239,0.002329936,0.009806726,0.8618242],"study_design_scores_gemma":[0.00002469877,0.000116149,0.008871322,0.00002352304,0.00004232174,0.0003900046,0.0001100818,0.9736782,0.009672527,0.002777977,0.004245755,0.00004738571],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1264278,0.0007294449,0.8360167,0.0006553657,0.0001182499,0.000277827,0.001643039,0.03274892,0.001382729],"genre_scores_gemma":[0.4556178,0.0002021261,0.5400087,0.0001916294,0.00004864875,0.0001407568,0.002145278,0.0001893548,0.001455656],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01157072,"threshold_uncertainty_score":0.02300674,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08194400033841359,"score_gpt":0.3697688114118219,"score_spread":0.2878248110734083,"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."}}