{"id":"W6964964439","doi":"10.25592/uhhfdm.1477","title":"Community Interpreting Database Pilot Corpus (ComInDat)","year":2010,"lang":"en","type":"dataset","venue":"Universität Hamburg","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"German; Metadata; Corpus linguistics; Speech community; Sample (material); Computational linguistics","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":[0.003221192,0.00210145,0.0016207,0.006102801,0.003012696,0.003681116,0.003307409,0.003167049,0.08734879],"category_scores_gemma":[0.01140298,0.001270828,0.0009902894,0.007471247,0.001400733,0.00374772,0.0050608,0.002776654,0.0504056],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002599087,"about_ca_system_score_gemma":0.003572406,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01561755,"about_ca_topic_score_gemma":0.02530515,"domain_scores_codex":[0.9965333,0.001162568,0.0006368637,0.0008656877,0.0006153187,0.0001862703],"domain_scores_gemma":[0.9917758,0.003638649,0.0004367143,0.001725047,0.00199554,0.000428193],"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":[0.0005763556,0.0002922812,0.001171857,0.00656295,0.00008934191,0.0009594258,0.002931213,0.000643858,0.005627901,0.005523178,0.9233227,0.05229891],"study_design_scores_gemma":[0.0005678835,0.00008967741,0.009307171,0.0007103572,0.00008691999,0.001408427,0.002248774,0.002020818,0.003727519,0.002144496,0.9775419,0.0001460755],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.01612056,0.00156466,0.008603025,0.0008400651,0.0004799176,0.001469142,0.9357044,0.006279253,0.02893912],"genre_scores_gemma":[0.01577205,0.000243801,0.01792149,0.0002006842,0.00005389465,0.003395947,0.9563481,0.001454038,0.004610116],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.08734879,"threshold_uncertainty_score":0.2922108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03492674166648898,"score_gpt":0.2679491332882719,"score_spread":0.2330223916217829,"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."}}