{"id":"W2619061592","doi":"","title":"Rarity and the problem of measuring diversity","year":2012,"lang":"en","type":"preprint","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Gut microbiota and health","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Taxon; Diversity (politics); Estimator; Sample (material); Metagenomics; Community structure; Range (aeronautics); Set (abstract data type); Sampling (signal processing); Computer science; Ecology; Geography; Statistics; Mathematics; Biology; Engineering","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":[0.03411273,0.001381629,0.002607104,0.009792851,0.00236114,0.005791333,0.003679792,0.003575244,0.00145766],"category_scores_gemma":[0.1772561,0.001344982,0.001398124,0.00714054,0.01763703,0.0114592,0.009150118,0.005125635,0.0006145015],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002446962,"about_ca_system_score_gemma":0.001089326,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001841077,"about_ca_topic_score_gemma":0.0008528055,"domain_scores_codex":[0.9441087,0.03418569,0.002703676,0.008499801,0.009373119,0.001129005],"domain_scores_gemma":[0.8295183,0.1294022,0.0146929,0.01812893,0.006387313,0.001870258],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002375369,0.00009494103,0.06592535,0.001315562,0.0007541471,0.0004750016,0.006040206,0.05170215,0.005751505,0.6129095,0.002800076,0.251994],"study_design_scores_gemma":[0.00002410743,0.00009265601,0.01378781,0.0002753425,0.00004197764,0.0006998076,0.0008975387,0.05279666,0.001976387,0.9223267,0.006942772,0.0001381785],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1701169,0.02284878,0.7785097,0.009040581,0.0004907436,0.0001535951,0.000683963,0.0003291277,0.01782654],"genre_scores_gemma":[0.8078802,0.004930175,0.1813144,0.001137235,0.001222571,0.0004961104,0.0004604385,0.0001508524,0.002408004],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03411273,"threshold_uncertainty_score":0.1804075,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01816936664434366,"score_gpt":0.2199892155980876,"score_spread":0.2018198489537439,"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."}}