{"id":"W7039306154","doi":"","title":"ManDi: Mandarin Chinese Dialect Corpus","year":2022,"lang":"en","type":"other","venue":"OSF Preprints (OSF Preprints)","topic":"Essential Oils and Antimicrobial Activity","field":"Agricultural and Biological Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Mandarin Chinese; Standard Chinese; Corpus linguistics; Register (sociolinguistics)","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":[0.001415306,0.001348531,0.0008287513,0.002840761,0.001616462,0.001276176,0.001401251,0.000693901,0.02712005],"category_scores_gemma":[0.003390222,0.0003888944,0.0003671229,0.003152386,0.0006748842,0.0009095125,0.001268073,0.000829756,0.01280064],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00138052,"about_ca_system_score_gemma":0.00310952,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03616399,"about_ca_topic_score_gemma":0.04247895,"domain_scores_codex":[0.9990118,0.0002341677,0.0001892855,0.0002827968,0.0001888982,0.00009317559],"domain_scores_gemma":[0.9985313,0.0003755022,0.00007061844,0.0002693513,0.000631071,0.0001220799],"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":[0.001216792,0.0003301535,0.01247506,0.005236871,0.000175561,0.002828218,0.004248327,0.001545513,0.04784421,0.01305075,0.7428905,0.168158],"study_design_scores_gemma":[0.0002696169,0.00009617147,0.06116094,0.0001783317,0.0001076311,0.001152895,0.001053469,0.003191414,0.0107041,0.00209312,0.9198539,0.0001385991],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.08692994,0.003178376,0.01340863,0.0008942092,0.0007077974,0.001888947,0.8527053,0.00377287,0.03651392],"genre_scores_gemma":[0.06601145,0.0006838772,0.02353145,0.0003468063,0.0001580825,0.004151261,0.8921914,0.0008065508,0.01211911],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03616399,"threshold_uncertainty_score":0.0907256,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009168311908261657,"score_gpt":0.2208215199935445,"score_spread":0.2116532080852828,"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."}}