{"id":"W4393789404","doi":"10.5281/zenodo.1456564","title":"Corpus Of Occitan Written Traditional Folktales Annotated With Part-Of-Speech (Owt-Tag)","year":2018,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Medieval European Literature and History","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"European Commission","keywords":"Natural language processing; Linguistics; Computer science; Philosophy","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":[0.0009535281,0.001266671,0.0008413707,0.005272679,0.002628172,0.001863452,0.0009851749,0.0006883483,0.05271614],"category_scores_gemma":[0.003080337,0.0005344088,0.0003917842,0.005345337,0.001221435,0.001165236,0.002745643,0.0008097686,0.02803222],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001757042,"about_ca_system_score_gemma":0.002378799,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01168391,"about_ca_topic_score_gemma":0.02651593,"domain_scores_codex":[0.9990821,0.0002227915,0.0001306873,0.0002704755,0.000193098,0.0001009425],"domain_scores_gemma":[0.9978834,0.0008700003,0.0001400211,0.000435168,0.0005281063,0.0001433579],"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":[0.002698892,0.0004467866,0.008710501,0.009841031,0.0001191151,0.004065896,0.01250344,0.001195195,0.04498329,0.01456674,0.6092764,0.2915927],"study_design_scores_gemma":[0.0001356924,0.00007662446,0.03323566,0.0006769308,0.00005905498,0.001162715,0.003040326,0.0005963949,0.006985505,0.00142202,0.9525208,0.00008821145],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1547028,0.005038211,0.0114084,0.0009861653,0.001203202,0.0009981249,0.6628367,0.006615408,0.156211],"genre_scores_gemma":[0.1136831,0.001212477,0.02015023,0.0002201636,0.0002540552,0.001923537,0.8267923,0.002389485,0.0333747],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05271614,"threshold_uncertainty_score":0.176353,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05822606488989326,"score_gpt":0.2203912336234934,"score_spread":0.1621651687336001,"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."}}