{"id":"W3158640552","doi":"10.2196/27976","title":"Public Discussion of Anthrax on Twitter: Using Machine Learning to Identify Relevant Topics and Events","year":2021,"lang":"en","type":"article","venue":"JMIR Public Health and Surveillance","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Categorization; Social media; Relevance (law); Bacillus anthracis; Data science; Information retrieval; Event (particle physics); Computer science; Artificial intelligence; Proxy (statistics); Machine learning; Natural language processing; World Wide Web; Biology","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001076303,0.000215259,0.0006445437,0.0002015317,0.0002064095,0.00006704436,0.00009637808,0.00008407463,0.00005547311],"category_scores_gemma":[0.001338019,0.0001611323,0.00006326614,0.0005363672,0.00006859055,0.0001568046,0.0002073754,0.0003032985,0.000006407802],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001147908,"about_ca_system_score_gemma":0.0006328344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006966476,"about_ca_topic_score_gemma":0.0001329331,"domain_scores_codex":[0.9973643,0.0003994598,0.0005622255,0.0005747176,0.0004656674,0.0006335839],"domain_scores_gemma":[0.9977362,0.000106751,0.0002219742,0.0004704823,0.0002458772,0.001218726],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000110348,0.0002113195,0.9475843,0.0007073112,0.00003359774,0.00003890143,0.0004496528,0.000002417386,0.0005397406,0.0001138708,0.0006907043,0.04951779],"study_design_scores_gemma":[0.001443107,0.0004047687,0.8471512,0.0002045284,0.000001901371,0.00009064916,0.0003211332,0.001404557,0.00001930268,0.0000493864,0.1486865,0.0002229594],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9518315,0.001659962,0.0002324214,0.04513581,0.0002209532,0.0005031239,0.0001184234,0.0000779435,0.0002198224],"genre_scores_gemma":[0.9942541,0.0008312293,0.0006676575,0.003001376,0.0001390995,0.00001894463,0.0003401297,0.00003195309,0.000715497],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1479958,"threshold_uncertainty_score":0.6570784,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07141352321373394,"score_gpt":0.3733272626778466,"score_spread":0.3019137394641127,"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."}}