{"id":"W6921540341","doi":"10.7939/dvn/10485","title":"TSM Corpus - A Version","year":2015,"lang":"en","type":"dataset","venue":"Borealis","topic":"Education Methods and Technologies","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Corpus linguistics; China; Computational linguistics; Text corpus","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":[0.002429536,0.001823869,0.001056839,0.005324432,0.001557639,0.002078781,0.003237456,0.001689266,0.03826008],"category_scores_gemma":[0.006241197,0.0007360069,0.0009106811,0.008526048,0.0007117541,0.00201361,0.002562034,0.002329785,0.05724398],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001868175,"about_ca_system_score_gemma":0.004216751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03305895,"about_ca_topic_score_gemma":0.05813676,"domain_scores_codex":[0.9981031,0.0004048582,0.0003810852,0.0004467954,0.0004425815,0.0002214333],"domain_scores_gemma":[0.9974838,0.0005455349,0.0001922788,0.0007547932,0.0007711798,0.0002523733],"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":[0.0001066502,0.00006122345,0.0006130034,0.000712315,0.00002288408,0.00009559899,0.0002258281,0.0002485869,0.0005777857,0.001411749,0.989122,0.006802343],"study_design_scores_gemma":[0.000135009,0.00001812675,0.005320426,0.0001765822,0.00002038084,0.0001341378,0.0002942398,0.0003195581,0.0007223916,0.0009023094,0.9919241,0.0000328258],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001112463,0.0001196521,0.000566036,0.0001007893,0.00007470318,0.0001302555,0.9941631,0.0008863294,0.002846713],"genre_scores_gemma":[0.0006138925,0.00003529074,0.0005922695,0.00002825858,0.000006531618,0.0002863826,0.9971861,0.0001265943,0.001124592],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03826008,"threshold_uncertainty_score":0.1279927,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07565520443162371,"score_gpt":0.4042287287701127,"score_spread":0.328573524338489,"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."}}