{"id":"W4376274788","doi":"10.1016/j.prevetmed.2023.105932","title":"Animal disease surveillance: How to represent textual data for classifying epidemiological information","year":2023,"lang":"en","type":"article","venue":"Preventive Veterinary Medicine","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke","funders":"Centre de Coopération Internationale en Recherche Agronomique pour le Développement; Agence Nationale de la Recherche; European Commission","keywords":"Relevance (law); Context (archaeology); Computer science; Representation (politics); Word (group theory); Natural language processing; Artificial intelligence; Binary classification; Information retrieval; Data science; Mathematics; Support vector machine; Geography","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":["metaresearch","metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002728836,0.0003256852,0.0007835315,0.0003183666,0.0001448769,0.00002598775,0.0005696227,0.00008663925,0.000162451],"category_scores_gemma":[0.01546637,0.0002491151,0.0001327398,0.0007083366,0.0002120879,0.0005735183,0.001047309,0.000204151,0.0001977805],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001024976,"about_ca_system_score_gemma":0.0001279152,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001618447,"about_ca_topic_score_gemma":0.000002717835,"domain_scores_codex":[0.9969995,0.0003898593,0.0006418321,0.0007908619,0.0005539419,0.0006239868],"domain_scores_gemma":[0.9959987,0.001116382,0.0002329866,0.00162564,0.0002577386,0.0007685189],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.01313885,0.0002868616,0.1253503,0.001781133,0.0003920511,0.000701997,0.0003698587,0.00003841586,0.00745323,0.0003808234,0.8115448,0.03856171],"study_design_scores_gemma":[0.002223896,0.00226379,0.7095025,0.0004189428,0.00008783378,0.00004422248,0.0004226262,0.006605196,0.000008764945,0.0001158709,0.2780501,0.0002562676],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8656824,0.0007915061,0.02084043,0.09359392,0.001133707,0.006253093,0.009124761,0.001404463,0.001175724],"genre_scores_gemma":[0.9758652,0.0002006995,0.001558766,0.001458477,0.0008532492,0.0003858037,0.01901205,0.00003397534,0.0006317313],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5841522,"threshold_uncertainty_score":0.9999961,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1931269320748563,"score_gpt":0.419078500008632,"score_spread":0.2259515679337757,"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."}}