{"id":"W4408613811","doi":"10.2196/54803","title":"Synthetic Data-Driven Approaches for Chinese Medical Abstract Sentence Classification: Computational Study","year":2025,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Traditional Chinese Medicine Studies","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Cluster analysis; Artificial intelligence; Sentence; Robustness (evolution); Annotation; Natural language processing; Machine learning; Information retrieval; Data mining","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"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":[0.005189592,0.001233985,0.0008882945,0.001897635,0.0007175346,0.0009634863,0.001730747,0.001271624,0.001688372],"category_scores_gemma":[0.01572183,0.0003804109,0.001481331,0.002015793,0.000731257,0.001257208,0.0008851443,0.001801166,0.0004038949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003168608,"about_ca_system_score_gemma":0.00174646,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01733167,"about_ca_topic_score_gemma":0.02014187,"domain_scores_codex":[0.9982142,0.0009512611,0.0001507007,0.0003737237,0.0002087478,0.0001014731],"domain_scores_gemma":[0.9760014,0.02009104,0.0009092946,0.0009516126,0.001737153,0.0003094879],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009310945,0.00107648,0.01814789,0.001019704,0.0003324712,0.0005343948,0.0004788418,0.7945655,0.002875231,0.00536054,0.01655126,0.1581265],"study_design_scores_gemma":[0.00002629449,0.00005015174,0.0008983215,0.000009861262,0.00001345881,0.00002567835,0.00004838917,0.9966384,0.0005861631,0.001242069,0.0004523124,0.000008844941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8234764,0.003125313,0.1558578,0.003099216,0.00043642,0.0007730744,0.007398657,0.002080651,0.003752301],"genre_scores_gemma":[0.8595518,0.0004935244,0.1233077,0.0004302858,0.0001981141,0.0006287729,0.01373889,0.00008486906,0.001565968],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01733167,"threshold_uncertainty_score":0.03446156,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2631650665508667,"score_gpt":0.490928070791608,"score_spread":0.2277630042407413,"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."}}