{"id":"W4396894334","doi":"10.1101/2024.05.10.593424","title":"Keys to the cabinet: unlocking biodiversity data in public entomology collections","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Species Distribution and Climate Change","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Northern British Columbia; Royal British Columbia Museum","funders":"","keywords":"Cabinet (room); Entomology; Biodiversity; Citizen science; Geography; Computer science; Data science; Database; Ecology; Biology; Archaeology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["open_science"],"consensus_categories":[],"category_scores_codex":[0.006137862,0.0004577276,0.0005444589,0.01531195,0.003494645,0.004678688,0.002149829,0.0006471469,0.05816178],"category_scores_gemma":[0.02056334,0.0009379413,0.0002326882,0.02098874,0.001191569,0.005386092,0.004795326,0.00137041,0.02589615],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006278471,"about_ca_system_score_gemma":0.01707376,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.3561285,"about_ca_topic_score_gemma":0.4957707,"domain_scores_codex":[0.9958954,0.0005635205,0.000444433,0.0004679007,0.002138828,0.0004898766],"domain_scores_gemma":[0.9783083,0.002567512,0.002287394,0.005643732,0.008099502,0.003093481],"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":[0.0001174141,0.00003515118,0.01077465,0.0004106591,0.00001022728,0.0001165422,0.002034382,0.0002292292,0.002543625,0.009512263,0.6804722,0.2937437],"study_design_scores_gemma":[0.00002360637,0.000009917571,0.02408787,0.0003795713,0.000008557091,0.00008835069,0.001138177,0.0004925044,0.001319607,0.001542661,0.9708629,0.00004621867],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.04354243,0.003901745,0.08328879,0.01062669,0.001563191,0.005617228,0.4921616,0.02764041,0.3316579],"genre_scores_gemma":[0.1277666,0.00459994,0.4253196,0.001223808,0.0004803401,0.004210318,0.330853,0.009090412,0.09645596],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9978502,"threshold_uncertainty_score":0.7081113,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04827269946343021,"score_gpt":0.2426128989499859,"score_spread":0.1943401994865557,"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."}}