{"id":"W3044693911","doi":"10.1017/s1351324920000509","title":"Comparison of rule-based and neural network models for negation detection in radiology reports","year":2020,"lang":"en","type":"article","venue":"Natural Language Engineering","topic":"Topic Modeling","field":"Computer Science","cited_by":43,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute of Population and Public Health","funders":"Engineering and Physical Sciences Research Council; Medical Research Council","keywords":"Computer science; Artificial intelligence; Natural language processing; Negation; Artificial neural network; Sentence; Syntax; Pipeline (software); Machine learning; Rule-based system; Python (programming language); Information retrieval; Programming language","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00009750339,0.00006617383,0.0001442086,0.00004260242,0.00001433579,0.00001341668,0.00008148398,0.00004761232,2.347706e-7],"category_scores_gemma":[0.00007415071,0.00006646637,0.00002367916,0.0001368261,0.00000402604,0.0001520727,0.00002889189,0.0001272434,3.090554e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001769227,"about_ca_system_score_gemma":0.000006589548,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001740674,"about_ca_topic_score_gemma":0.000006143766,"domain_scores_codex":[0.9994282,0.00001166156,0.0001987108,0.0001731147,0.00006151239,0.0001267684],"domain_scores_gemma":[0.9997228,0.00007372344,0.00005840383,0.00009897994,0.00001686244,0.00002928113],"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":[0.000005816586,0.000001793094,0.00026882,0.00003372312,0.000001988279,0.000005468722,0.0007470753,0.9632947,0.01039086,0.0001993122,0.00000182118,0.02504862],"study_design_scores_gemma":[0.0001637632,0.00002628613,0.0004410198,0.00001200241,0.000001978199,0.000007529035,0.00001746967,0.9926206,0.006583218,0.00005779203,0.000004989408,0.00006336585],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3909537,0.001289456,0.6073574,0.00007720487,0.0001790095,0.00008836335,1.967373e-7,0.00005231012,0.000002268392],"genre_scores_gemma":[0.9302409,5.636725e-7,0.06959669,0.00005182119,0.00009411179,0.000007472136,0.000002671831,0.000005001656,7.32634e-7],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5392872,"threshold_uncertainty_score":0.271042,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01456388340755196,"score_gpt":0.2477711294140073,"score_spread":0.2332072460064553,"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."}}