{"id":"W4253082290","doi":"10.1037/e527382013-003","title":"High Quality Analytics with Poor Quality Data","year":2012,"lang":"en","type":"dataset","venue":"PsycEXTRA Dataset","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Northern British Columbia","funders":"","keywords":"Quality (philosophy); Analytics; Data quality; Computer science; Data science; Data analysis; Data mining; Business; Marketing","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":[0.008672291,0.001180584,0.001103411,0.004492563,0.001203773,0.006223937,0.002051149,0.001325308,0.001695807],"category_scores_gemma":[0.03410176,0.0005817708,0.0009112686,0.01095485,0.0009455403,0.004421049,0.003830362,0.002395673,0.001768602],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001414612,"about_ca_system_score_gemma":0.00222848,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005304712,"about_ca_topic_score_gemma":0.005994741,"domain_scores_codex":[0.9806218,0.002527771,0.002675068,0.002708945,0.01058971,0.0008767049],"domain_scores_gemma":[0.9562855,0.01068354,0.003979099,0.01759434,0.01071511,0.000742343],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001928398,0.001368809,0.2223876,0.003680008,0.000980387,0.002239216,0.002091265,0.04191703,0.05069466,0.0247848,0.2393539,0.4085739],"study_design_scores_gemma":[0.0003950362,0.0005025513,0.1425461,0.0007012398,0.0004804482,0.002598876,0.002656964,0.2305445,0.08762038,0.06745306,0.46419,0.0003109124],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"dataset","genre_scores_codex":[0.3573522,0.003892708,0.3598458,0.008983837,0.0009490283,0.001987822,0.2286506,0.01925456,0.01908348],"genre_scores_gemma":[0.3766329,0.001485157,0.2676959,0.001041176,0.0003028559,0.001015252,0.3473228,0.000716158,0.003787867],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.008672291,"threshold_uncertainty_score":0.04586405,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1232503504945862,"score_gpt":0.3961428459929356,"score_spread":0.2728924954983494,"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."}}