{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002117968,0.00007950013,0.0001452864,0.0001325811,0.0001938142,0.000597074,0.0007927285,0.0000659611,0.000001438209],"category_scores_gemma":[0.00008071769,0.00006465517,0.00003261184,0.0002720591,0.00001384393,0.0131519,0.0003088981,0.0001405456,0.000002657169],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007920488,"about_ca_system_score_gemma":0.00006613287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002994589,"about_ca_topic_score_gemma":0.000001940222,"domain_scores_codex":[0.9986984,0.00004213219,0.000817357,0.0001578247,0.0001782967,0.0001060108],"domain_scores_gemma":[0.9985504,0.0001371883,0.0004268601,0.0006579789,0.0001611736,0.00006639721],"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":[0.00003397198,0.00003673903,0.00002104803,0.0006178017,0.00008695166,0.000001057337,0.0006250154,0.000009430137,0.0007996479,0.0299416,0.07652248,0.8913043],"study_design_scores_gemma":[0.0001191445,0.0000619399,0.00004982113,0.0000351631,0.00001558867,0.0001584316,0.000179155,0.7250628,0.0005627057,0.0003236596,0.273357,0.00007456137],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001213028,0.0002333891,0.9966207,0.000775709,0.000218744,0.0001719187,0.0003239151,0.000325828,0.0001167493],"genre_scores_gemma":[0.9201317,0.000182045,0.07889946,0.0002109844,0.0001114307,0.00001176138,0.0004360392,0.000005614565,0.00001098779],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9189187,"threshold_uncertainty_score":0.9534814,"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."}}