{"id":"W3022964795","doi":"10.2196/17637","title":"Multi-Level Representation Learning for Chinese Medical Entity Recognition: Model Development and Validation","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China","keywords":"Computer science; Natural language processing; Artificial intelligence; Softmax function; Language model; Deep learning; Feature learning; Named-entity recognition; Representation (politics); Encoder; Information retrieval; Task (project management)","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":[0.005577537,0.001417827,0.001139905,0.001198854,0.0006259325,0.001123637,0.002195215,0.001529928,0.002698984],"category_scores_gemma":[0.007354037,0.0005021294,0.001635197,0.001265619,0.0004932365,0.002046138,0.001581883,0.003297642,0.0009497522],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002436914,"about_ca_system_score_gemma":0.002498371,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04269174,"about_ca_topic_score_gemma":0.01994521,"domain_scores_codex":[0.9989458,0.0003804593,0.0000817806,0.0003077066,0.0001536772,0.0001306004],"domain_scores_gemma":[0.9957817,0.002518191,0.0002105321,0.0003389564,0.001062119,0.00008849554],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0004400722,0.0004903282,0.006219201,0.000213752,0.000250019,0.0001276109,0.0001142442,0.7242178,0.001826439,0.002164947,0.004205841,0.2597298],"study_design_scores_gemma":[0.000008949337,0.00003250777,0.0003009408,0.00000754544,0.00001541479,0.000007650074,0.00000898188,0.9984477,0.0005883901,0.0004209377,0.0001546126,0.000006326344],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3732184,0.00520963,0.605723,0.002100035,0.0003337454,0.0008288086,0.002082617,0.005352117,0.005151726],"genre_scores_gemma":[0.8453056,0.001008825,0.1457435,0.0003088813,0.00006674422,0.0007855954,0.003519764,0.0001117203,0.003149366],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04269174,"threshold_uncertainty_score":0.08488649,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1171216017315767,"score_gpt":0.3830128472133887,"score_spread":0.265891245481812,"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."}}