{"id":"W4280635309","doi":"10.1371/journal.pone.0268426","title":"Addressing the need for interactive, efficient, and reproducible data processing in ecology with the datacleanr R package","year":2022,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Data Analysis with R","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université TÉLUQ; Université du Québec à Montréal","funders":"Helmholtz-Gemeinschaft; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; National Science Foundation","keywords":"Computer science; Workflow; Data science; Data quality; Data processing; Outlier; Data type; Data mining; Quality (philosophy); Database; Service (business); Artificial intelligence; Programming language","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":[0.05281568,0.003436408,0.003681189,0.004425373,0.002311932,0.009618538,0.007069865,0.002576547,0.01648036],"category_scores_gemma":[0.1367923,0.00343038,0.005407551,0.005763336,0.004262418,0.006368819,0.01220844,0.009922777,0.01816745],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001218494,"about_ca_system_score_gemma":0.01035304,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004285141,"about_ca_topic_score_gemma":0.006895307,"domain_scores_codex":[0.9642435,0.02169995,0.002715108,0.004246215,0.006233248,0.0008619201],"domain_scores_gemma":[0.8659194,0.09437052,0.007321288,0.02137463,0.008988289,0.002025926],"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":[0.001005844,0.0003311649,0.0230681,0.009005305,0.003537775,0.001413228,0.006632223,0.02735789,0.02795359,0.07738071,0.5792705,0.2430436],"study_design_scores_gemma":[0.0007342305,0.0003241725,0.01290686,0.001836469,0.0009238549,0.001460121,0.0007212068,0.08533382,0.03470796,0.2017375,0.6582787,0.001035067],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.003569574,0.00051097,0.8787217,0.002906265,0.0003565035,0.000345553,0.01150782,0.09996682,0.002114898],"genre_scores_gemma":[0.01822723,0.0004937322,0.9153088,0.002003551,0.000202022,0.002646611,0.01108496,0.04882465,0.001208398],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.05281568,"threshold_uncertainty_score":0.2793193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1763984246597676,"score_gpt":0.3123639623910735,"score_spread":0.1359655377313059,"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."}}