{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01214513,0.0007926505,0.0009472732,0.001012698,0.000743271,0.0009795228,0.002222788,0.001308733,0.0030956],"category_scores_gemma":[0.0273665,0.0006812893,0.001189988,0.0009238489,0.0006938018,0.001146769,0.001499332,0.00216184,0.0006042271],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006383102,"about_ca_system_score_gemma":0.001770017,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002715035,"about_ca_topic_score_gemma":0.003342635,"domain_scores_codex":[0.9974803,0.001701611,0.0001488311,0.0003008997,0.0002404316,0.0001278066],"domain_scores_gemma":[0.98574,0.01057401,0.0007767276,0.001114722,0.001479977,0.0003145278],"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":[0.0004120275,0.000164495,0.008704441,0.000139583,0.0002235873,0.0001896727,0.0002442531,0.9255714,0.002736371,0.03200421,0.001576185,0.02803377],"study_design_scores_gemma":[0.00003592357,0.00003726377,0.000150278,0.000009282676,0.0000137396,0.00001978828,0.000009782679,0.9943547,0.0007500522,0.004066794,0.0005415746,0.00001072885],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01730871,0.00002584384,0.9814118,0.0001015468,0.00002423606,0.00018575,0.0001237877,0.0004553684,0.0003629997],"genre_scores_gemma":[0.1833745,0.0000428087,0.8135775,0.0001383758,0.00002479847,0.001431675,0.0006604607,0.000171966,0.0005779055],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01214513,"threshold_uncertainty_score":0.06423038,"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."}}