{"id":"W2797647106","doi":"10.1145/3190577","title":"InfoClean","year":2017,"lang":"en","type":"article","venue":"Journal of Data and Information Quality","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Computer science; Confidentiality; Information sensitivity; Data mining; Process (computing); Data quality; Set (abstract data type); Information privacy; Information loss; Data set; Database; Data science; Computer security; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00358904,0.002001266,0.001758306,0.005443807,0.002379751,0.008001858,0.004146279,0.0024005,0.0660184],"category_scores_gemma":[0.01714829,0.001316341,0.003210411,0.005452708,0.00108202,0.008340333,0.006678186,0.003096503,0.05080087],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001226624,"about_ca_system_score_gemma":0.003861419,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003355052,"about_ca_topic_score_gemma":0.004581771,"domain_scores_codex":[0.9954797,0.0006764446,0.0004586148,0.00109102,0.001986595,0.0003076331],"domain_scores_gemma":[0.9908944,0.002087051,0.0004648444,0.004645258,0.001661191,0.0002472372],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000633198,0.0002024184,0.004495177,0.001681167,0.0003421363,0.000476353,0.0005889387,0.004188647,0.004322792,0.04188948,0.5373911,0.4037886],"study_design_scores_gemma":[0.00007639844,0.00007455832,0.0013501,0.0002446245,0.00008516922,0.0005479817,0.0002700013,0.01819371,0.01437706,0.04117215,0.9235305,0.00007787591],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007545008,0.004687097,0.5485691,0.004734895,0.001684959,0.001084138,0.07921217,0.2595254,0.09295721],"genre_scores_gemma":[0.07170301,0.004382,0.5481141,0.005576994,0.0006767907,0.001135921,0.259783,0.03945182,0.06917652],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0660184,"threshold_uncertainty_score":0,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1247884661779293,"score_gpt":0.3855261588859472,"score_spread":0.2607376927080179,"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."}}