{"id":"W3151875318","doi":"10.2196/23305","title":"Effective Training Data Extraction Method to Improve Influenza Outbreak Prediction from Online News Articles: Deep Learning Model Study","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Research Foundation of Korea; National Research Foundation","keywords":"Computer science; Word embedding; Machine learning; Artificial intelligence; Microblogging; Social media; Data mining; Word (group theory); Embedding; World Wide Web","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":[0.001844277,0.001215878,0.001018021,0.001275477,0.0004274609,0.00076754,0.001173004,0.001274554,0.001030372],"category_scores_gemma":[0.004630711,0.0004400786,0.00097164,0.0009031951,0.0003428416,0.001565472,0.000587676,0.001824575,0.0003415227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009774938,"about_ca_system_score_gemma":0.001039054,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01304329,"about_ca_topic_score_gemma":0.009958218,"domain_scores_codex":[0.9995523,0.0001062323,0.00004809832,0.0001394427,0.00007707722,0.00007678961],"domain_scores_gemma":[0.9970683,0.00185483,0.0001631427,0.0001065989,0.0007274307,0.00007959349],"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":[0.0003708652,0.001257096,0.01707214,0.000347148,0.0002939957,0.0003529745,0.0002455209,0.6234098,0.00595289,0.002619193,0.005698472,0.3423799],"study_design_scores_gemma":[0.000003036356,0.00001878968,0.0002523469,0.000004078299,0.00001077662,0.000006856558,0.000006359939,0.9990226,0.0004252623,0.0001746788,0.00007293122,0.0000022489],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4548513,0.005215534,0.5310214,0.002217432,0.0003722906,0.0002492676,0.0005311206,0.001545419,0.003996245],"genre_scores_gemma":[0.9159669,0.001156427,0.07760409,0.000322181,0.0001641431,0.0001767951,0.0008926457,0.00004966612,0.003667209],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01304329,"threshold_uncertainty_score":0.02593476,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05695722559282902,"score_gpt":0.4006417169882084,"score_spread":0.3436844913953794,"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."}}