{"id":"W6976873916","doi":"10.6084/m9.figshare.20539056.v2","title":"File 3. Taxonomy and Quality Counts","year":2023,"lang":"en","type":"dataset","venue":"Figshare","topic":"Aquatic Ecosystems and Biodiversity","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Taxonomy (biology); Completeness (order theory); Quality (philosophy); Data quality","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002037594,0.001963973,0.001858257,0.006132379,0.001441999,0.003284859,0.002989074,0.001999928,0.4512354],"category_scores_gemma":[0.01546757,0.001065934,0.001360383,0.01129291,0.0005530988,0.003053555,0.002516681,0.002297662,0.2342003],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002453106,"about_ca_system_score_gemma":0.003188957,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02620774,"about_ca_topic_score_gemma":0.04067355,"domain_scores_codex":[0.9983466,0.0001964278,0.0002983691,0.0005270601,0.0003714312,0.0002601377],"domain_scores_gemma":[0.9901587,0.003935513,0.0007834281,0.001455835,0.003157465,0.0005090057],"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":[0.00004432565,0.00001081976,0.0005121655,0.0006452543,0.00001691973,0.00001279547,0.00002516213,0.0001054701,0.00007125021,0.0003360379,0.9965791,0.00164066],"study_design_scores_gemma":[0.0003186964,0.0000184647,0.004196634,0.0006191,0.00003168386,0.0000562678,0.0001173953,0.0002197648,0.0002677719,0.002375348,0.9917402,0.00003861495],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00001762084,0.000005579343,0.00002963705,0.00001755174,0.00000628708,0.000004968176,0.999526,0.0001278254,0.000264499],"genre_scores_gemma":[0.000181782,0.000016278,0.0003106193,0.00004362659,0.000004781262,0.000063427,0.9986514,0.000180992,0.0005470855],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.4512354,"threshold_uncertainty_score":0.7827461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1110733128754445,"score_gpt":0.2554523747941808,"score_spread":0.1443790619187363,"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."}}