{"id":"W2990170152","doi":"10.2196/13347","title":"Lifestyle Disease Surveillance Using Population Search Behavior: Feasibility Study","year":2019,"lang":"en","type":"article","venue":"Journal of Medical Internet Research","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Overfitting; Generalizability theory; Multivariate statistics; Covariate; Computer science; Context (archaeology); Proxy (statistics); Population; Feature selection; Medicine; Statistics; Machine learning; Geography; Environmental health; Mathematics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.02068223,0.0008451154,0.0006711896,0.001641985,0.0007930195,0.0008930222,0.001426751,0.001380294,0.002378588],"category_scores_gemma":[0.03953682,0.0006346046,0.001254615,0.001312072,0.0008095966,0.002632008,0.001522462,0.001231432,0.001150621],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001302275,"about_ca_system_score_gemma":0.003499715,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01688581,"about_ca_topic_score_gemma":0.01127516,"domain_scores_codex":[0.9893495,0.008499778,0.0003895527,0.0006467582,0.0006353541,0.0004791658],"domain_scores_gemma":[0.9743634,0.01593237,0.002063314,0.002763722,0.00322903,0.001648155],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.005585179,0.03568231,0.8538421,0.0008727945,0.0005448366,0.0008846489,0.003623171,0.008697339,0.001415704,0.001368907,0.002672342,0.08481058],"study_design_scores_gemma":[0.003121003,0.08929428,0.7756477,0.0003086586,0.0006312281,0.001282051,0.007981896,0.1109588,0.002356477,0.002130929,0.006037412,0.0002495611],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.987963,0.00008473953,0.004914013,0.0003629318,0.00001982452,0.004167777,0.001029769,0.00005431824,0.001403655],"genre_scores_gemma":[0.9820133,0.0001428359,0.01098263,0.0002024587,0.00004591196,0.004742232,0.001473635,0.00001123032,0.0003857632],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02068223,"threshold_uncertainty_score":0.1093794,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1238580738224365,"score_gpt":0.4834293947039309,"score_spread":0.3595713208814943,"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."}}