{"id":"W4396561978","doi":"10.1201/9781003258254","title":"The Data Preparation Journey","year":2024,"lang":"en","type":"book","venue":"","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Computer science","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":[0.02675178,0.001346662,0.001560942,0.003264036,0.004849866,0.02240234,0.00426034,0.004777827,0.08475397],"category_scores_gemma":[0.1000951,0.002030041,0.001981545,0.004957607,0.004336218,0.01699919,0.01229948,0.0143083,0.1184379],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004464178,"about_ca_system_score_gemma":0.01512283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003048233,"about_ca_topic_score_gemma":0.003488309,"domain_scores_codex":[0.9733483,0.00897971,0.001660158,0.0027939,0.01246922,0.0007486868],"domain_scores_gemma":[0.9308346,0.03100936,0.001823875,0.01057856,0.02070048,0.005053097],"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":[0.00004168298,0.0000452134,0.000247798,0.0005137438,0.00002054301,0.0001603072,0.002028266,0.0003389405,0.0005192432,0.05152305,0.7159216,0.2286396],"study_design_scores_gemma":[0.000004120973,0.00001376056,0.00009220328,0.0002471303,0.000003337234,0.0001187369,0.0002970021,0.00009260229,0.0001705372,0.01251463,0.9864274,0.00001846128],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.003146675,0.01879496,0.365862,0.1796789,0.02919972,0.002626653,0.01197315,0.01635717,0.3723609],"genre_scores_gemma":[0.01072432,0.01963004,0.3928907,0.05237713,0.006934996,0.002934049,0.01334383,0.01493575,0.4862293],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.08475397,"threshold_uncertainty_score":0.2835302,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5389770882399062,"score_gpt":0.532252402529568,"score_spread":0.006724685710338241,"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."}}