{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006817764,0.001915414,0.001481442,0.008130921,0.002302694,0.002359254,0.003539357,0.001657786,0.03609045],"category_scores_gemma":[0.004459785,0.0005798377,0.00101609,0.01373089,0.0006846403,0.001025439,0.00231317,0.002191233,0.04457623],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005545605,"about_ca_system_score_gemma":0.009344406,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.6072004,"about_ca_topic_score_gemma":0.7716185,"domain_scores_codex":[0.9990575,0.00008525983,0.00008284237,0.0001841916,0.000328561,0.0002615465],"domain_scores_gemma":[0.9976314,0.0003363809,0.0001280395,0.0003487588,0.001277499,0.0002777932],"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":[0.00004197492,0.00002390507,0.001281549,0.0004996437,0.0000246009,0.00005310779,0.00009144134,0.0001516081,0.0001237221,0.0003930232,0.9928508,0.004464627],"study_design_scores_gemma":[0.00009648453,0.000009152897,0.01965921,0.0004648307,0.00003290497,0.00009926127,0.0004819387,0.0004266486,0.0003501242,0.0004642817,0.9778783,0.0000369732],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0004494265,0.00009414995,0.00003627153,0.00003972248,0.00001845038,0.0000109847,0.998351,0.000131983,0.0008680384],"genre_scores_gemma":[0.0006020793,0.00004663671,0.0001198911,0.00001221818,0.000002680146,0.00004791596,0.9984707,0.00001979114,0.0006780854],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.3927996,"threshold_uncertainty_score":0.7902257,"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."}}