{"id":"W6944952617","doi":"10.21415/t5nk63","title":"CHILDES French York Corpus","year":2004,"lang":"en","type":"dataset","venue":"TalkBank","topic":"Cell Image Analysis Techniques","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Corpus linguistics; Feature (linguistics); Context (archaeology); Subject (documents)","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":[0.001233351,0.002434555,0.00132366,0.008939399,0.00194037,0.00218466,0.001827734,0.001860246,0.04683461],"category_scores_gemma":[0.004883162,0.0005966275,0.001058679,0.008113983,0.0008156932,0.001282907,0.001686467,0.001439671,0.027738],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006199855,"about_ca_system_score_gemma":0.007940811,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.6531593,"about_ca_topic_score_gemma":0.7170843,"domain_scores_codex":[0.9986151,0.0002502228,0.0001014861,0.0004931648,0.0003132484,0.0002266643],"domain_scores_gemma":[0.9965876,0.001110551,0.000158942,0.0004618896,0.001416184,0.0002648892],"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":[0.0003084243,0.00005383872,0.002990558,0.0006118205,0.0001132952,0.0002338944,0.0003663801,0.0008568023,0.001010294,0.001754538,0.977037,0.01466319],"study_design_scores_gemma":[0.000222113,0.00002653483,0.04812972,0.0002660404,0.0001120101,0.0003373615,0.0007595949,0.001443479,0.001852749,0.0006523528,0.9461174,0.00008059474],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.008331631,0.0007681159,0.0003435988,0.0002445812,0.0000624366,0.0000384445,0.9831055,0.00090163,0.00620405],"genre_scores_gemma":[0.006927119,0.0001691423,0.0006417694,0.00004674788,0.00001443965,0.0001002492,0.9875203,0.0002112989,0.004368914],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.6531593,"threshold_uncertainty_score":0.6977667,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00608254786747934,"score_gpt":0.2465219452571228,"score_spread":0.2404393973896435,"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."}}