{"id":"W3186788573","doi":"10.35111/srvm-p675","title":"LORELEI Akan Representative Language Pack","year":2021,"lang":"en","type":"dataset","venue":"Linguistic Data Consortium Catalog","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Annotation; Natural language processing; XML; Artificial intelligence; Context (archaeology); Resource (disambiguation); Sentence; Lemmatisation; Information retrieval; World Wide Web","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003045092,0.001373664,0.000937717,0.005037313,0.001521139,0.00439326,0.002005257,0.0007898976,0.30194],"category_scores_gemma":[0.009225233,0.0009789636,0.000515877,0.005111583,0.0007246207,0.005877991,0.004186717,0.001689443,0.2748518],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001864814,"about_ca_system_score_gemma":0.002793196,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01326982,"about_ca_topic_score_gemma":0.01276855,"domain_scores_codex":[0.9972004,0.0006159556,0.0002963838,0.0006403611,0.001029043,0.0002179546],"domain_scores_gemma":[0.9888527,0.002393277,0.0005769805,0.002597003,0.00503947,0.0005403791],"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":[0.0002347462,0.0000449861,0.0007070818,0.0004402121,0.000009297602,0.0001877583,0.0006234452,0.0002212784,0.003292623,0.003243646,0.896374,0.09462098],"study_design_scores_gemma":[0.00001431512,0.0000158082,0.0009552281,0.00007740709,0.000003815332,0.0001357782,0.0002902258,0.0004418933,0.002056052,0.0007523025,0.995233,0.00002419712],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"dataset","genre_scores_codex":[0.01383552,0.001617411,0.0464054,0.00488804,0.001995478,0.001682039,0.3872791,0.08344302,0.458854],"genre_scores_gemma":[0.03299748,0.001199231,0.06673688,0.001504821,0.0006293324,0.002793233,0.5173878,0.0193678,0.3573834],"genre_candidate":"dataset","genre_consensus":null,"teacher_disagreement_score":0.30194,"threshold_uncertainty_score":0.9956979,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02960889486193685,"score_gpt":0.3335016712738956,"score_spread":0.3038927764119588,"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."}}