{"id":"W4353016808","doi":"10.1371/journal.pcbi.1010154","title":"A real data-driven simulation strategy to select an imputation method for mixed-type trait data","year":2023,"lang":"en","type":"article","venue":"PLoS Computational Biology","topic":"Amphibian and Reptile Biology","field":"Environmental Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"Natural Sciences and Engineering Research Council of Canada; Ontario Ministry of Economic Development, Job Creation and Trade; Canada First Research Excellence Fund; Government of Canada; Ministero dello Sviluppo Economico; Genome Canada; Ontario Genomics; University of Guelph","keywords":"Missing data; Imputation (statistics); Categorical variable; Random forest; Statistics; Trait; Type I and type II errors; Multivariate statistics; Computer science; Data mining; Mathematics; Artificial intelligence","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":[0.0004536946,0.0001382715,0.0001914181,0.00008530071,0.0001399333,0.00002038425,0.0006819087,0.0001363648,0.0002208332],"category_scores_gemma":[0.0001813394,0.000123861,0.0000219719,0.0004163962,0.00008342882,0.0002383509,0.0004753518,0.00007017949,0.0003801827],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005575194,"about_ca_system_score_gemma":0.00006021266,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001255448,"about_ca_topic_score_gemma":0.0002194248,"domain_scores_codex":[0.9982472,0.0002450404,0.0002800096,0.0008112863,0.0001246967,0.000291731],"domain_scores_gemma":[0.9985962,0.0007129177,0.00009545608,0.0004409882,0.0000419757,0.0001124676],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009040961,0.00006647096,0.00204265,0.000005816173,0.00004089938,0.000001425013,0.0001037049,0.9396716,0.0101559,0.0006278379,0.003622192,0.04357113],"study_design_scores_gemma":[0.0002623976,0.0005846264,0.03655529,0.000002093381,0.00002128354,0.000002997927,0.00003608666,0.9538516,0.00003228887,0.005974446,0.00252111,0.0001557171],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6066335,0.000005467032,0.3892666,0.00053552,0.0002805013,0.0008910979,0.001795585,0.0002695703,0.000322133],"genre_scores_gemma":[0.8446954,0.000003496686,0.1195327,0.0002620013,0.0001841644,0.00001979137,0.03521186,0.00001917262,0.00007138946],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.269734,"threshold_uncertainty_score":0.5050906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1246956428727454,"score_gpt":0.3916229891582338,"score_spread":0.2669273462854885,"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."}}