{"id":"W4320914218","doi":"10.36227/techrxiv.22060118","title":"DataCurator.jl: Efficient, portable, and reproducible validation, curation, and transformation of large heterogeneous datasets using human-readable recipes compiled into machine verifiable templates","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Simon Fraser University","funders":"","keywords":"Computer science; Executable; Scalability; Python (programming language); Workflow; Preprocessor; Verifiable secret sharing; Data curation; Scripting language; Data mining; Database; Programming language; Set (abstract data type)","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":["metaresearch","open_science"],"consensus_categories":[],"category_scores_codex":[0.01304276,0.003839812,0.002594315,0.003792744,0.001845173,0.005920354,0.006190197,0.00192289,0.04559891],"category_scores_gemma":[0.0347551,0.00488045,0.004050964,0.002709067,0.003135295,0.005509269,0.01017857,0.006511203,0.07788678],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001666262,"about_ca_system_score_gemma":0.006682278,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003839154,"about_ca_topic_score_gemma":0.005482656,"domain_scores_codex":[0.9916618,0.001285055,0.0009951208,0.002347506,0.003097205,0.0006132715],"domain_scores_gemma":[0.9786075,0.006940769,0.001869972,0.007946732,0.003698941,0.0009361347],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0009789374,0.0001986811,0.005443362,0.003120699,0.0005562418,0.0007219492,0.001518637,0.005403243,0.03667413,0.01423187,0.8002869,0.1308653],"study_design_scores_gemma":[0.0005503225,0.0002022202,0.005152238,0.000863531,0.0001655914,0.0006919031,0.0002490366,0.03949705,0.1085033,0.03258536,0.810888,0.0006514912],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.002005239,0.0002944705,0.281817,0.000496542,0.0003661772,0.0005644659,0.04219911,0.6677538,0.004503211],"genre_scores_gemma":[0.01733245,0.0006079119,0.4002195,0.001774049,0.0001549673,0.003690504,0.1405812,0.4249906,0.01064875],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.9938098,"threshold_uncertainty_score":0.1525436,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1992121274069104,"score_gpt":0.4319443258583797,"score_spread":0.2327321984514693,"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."}}