{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","research_integrity","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001512197,0.001053474,0.001018379,0.001277443,0.001306904,0.0002138598,0.004439819,0.0007547826,0.005819603],"category_scores_gemma":[0.0008481971,0.001243338,0.0002833051,0.0007295639,0.0008494748,0.001724795,0.00458244,0.009100013,0.01689573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008149769,"about_ca_system_score_gemma":0.0004965509,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.02781045,"about_ca_topic_score_gemma":0.02279448,"domain_scores_codex":[0.9954547,0.001290095,0.0005671831,0.0008652879,0.0007984125,0.001024296],"domain_scores_gemma":[0.9922437,0.000792027,0.0009358216,0.005113983,0.0003221669,0.000592322],"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":[0.000517892,0.0006329012,0.00002220241,0.0001647071,0.0002644245,0.0007238304,0.0001481771,0.00000278511,0.002215981,0.00005977241,0.9950952,0.000152162],"study_design_scores_gemma":[0.001434626,0.0003220381,0.00001927257,0.0005410244,0.0006435775,0.0001158333,0.0006123529,0.00003311503,0.0002098714,0.00003741694,0.9948245,0.001206354],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001963756,0.00006492279,0.00006608901,0.00006934539,0.001438132,0.0005856504,0.9926841,0.0004260587,0.002701972],"genre_scores_gemma":[0.0009173923,0.00006097503,0.0003854265,0.0003547551,0.000280816,0.000004572349,0.9967157,0.0001869752,0.001093357],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01107613,"threshold_uncertainty_score":0.9999933,"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."}}