{"id":"W4393415790","doi":"10.5281/zenodo.8085892","title":"CLDF dataset derived from Tolmie and Dawson's \"Comparative Vocabulary of the Indigenous Peoples in British Columbia\" from 1884","year":2023,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Linguistic Variation and Morphology","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Indigenous; Vocabulary; Geography; Genealogy; Anthropology; History; Ethnology; Linguistics; Sociology; Biology; Philosophy; Ecology","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["sts","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0006600585,0.0001188532,0.0003149562,0.00009761879,0.001859611,0.000959805,0.001384877,0.0002047309,0.006860819],"category_scores_gemma":[0.002771068,0.0001593531,0.00003841364,0.0004893679,0.0005189829,0.00009525567,0.001446513,0.0003996936,0.001534915],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001363251,"about_ca_system_score_gemma":0.00005922956,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.3425125,"about_ca_topic_score_gemma":0.1493925,"domain_scores_codex":[0.997229,0.001176494,0.0003419369,0.0004833772,0.0004542329,0.0003149412],"domain_scores_gemma":[0.998628,0.0002400334,0.000264396,0.0004873767,0.0002550972,0.0001251473],"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":[0.00001192764,0.00006630315,0.00008093398,0.00002060365,0.00005183991,0.00003823795,0.006939539,0.000001833214,0.00003847535,0.00001796605,0.9918174,0.0009149539],"study_design_scores_gemma":[0.0003899822,0.00003916014,0.02076309,0.00008153466,0.00003722853,0.000005278233,0.003826717,0.000004197146,0.000004263618,0.0001909876,0.9744977,0.0001598074],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.04358349,0.0001083924,0.00001031894,0.0001674134,0.0003261258,0.0005352751,0.9545909,0.00009264086,0.0005854246],"genre_scores_gemma":[0.01479623,0.0008245087,0.00004277499,0.0001585722,0.000256937,1.505993e-7,0.9835026,0.0002236741,0.0001945275],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.19312,"threshold_uncertainty_score":0.9994398,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04476413154644514,"score_gpt":0.2878193656891803,"score_spread":0.2430552341427351,"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."}}