{"id":"W4232504079","doi":"10.2196/preprints.13347","title":"Lifestyle Disease Surveillance Using Population Search Behavior: Feasibility Study (Preprint)","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Overfitting; Generalizability theory; Covariate; Multivariate statistics; Computer science; Population; Proxy (statistics); Baseline (sea); Statistics; Medicine; Econometrics; Artificial intelligence; Machine learning; 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":[{"model":"gemma","categories":[],"domain":null,"study_design":"observational","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"design_other","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"high","status":"direct model label, unvalidated"}],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02097812,0.000557858,0.000494287,0.0009793548,0.0008397087,0.0009585164,0.001101636,0.001107516,0.005525852],"category_scores_gemma":[0.04918781,0.0004765034,0.001047771,0.001024406,0.0005717953,0.002137308,0.001232833,0.0009982958,0.002865128],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001016959,"about_ca_system_score_gemma":0.004268269,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01658996,"about_ca_topic_score_gemma":0.009655537,"domain_scores_codex":[0.9927395,0.005573468,0.0004046137,0.0004370987,0.0005079329,0.000337433],"domain_scores_gemma":[0.9639794,0.02454865,0.002130761,0.003201909,0.004026047,0.002113225],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.007658719,0.045661,0.7822223,0.001695761,0.0004837326,0.000858901,0.006186323,0.003680895,0.001600571,0.001395383,0.01797737,0.130579],"study_design_scores_gemma":[0.004746019,0.09052797,0.8517613,0.0004169417,0.000527933,0.0006400847,0.01097809,0.02547787,0.002462712,0.00161476,0.01061639,0.0002299385],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9789076,0.0001017804,0.004123009,0.001217352,0.00007859699,0.008311181,0.004469004,0.0001454696,0.002645914],"genre_scores_gemma":[0.9609681,0.0002307842,0.01557364,0.0008660803,0.0002096548,0.01584547,0.004815767,0.00003730559,0.001453124],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02097812,"threshold_uncertainty_score":0.1109443,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07353565567406418,"score_gpt":0.3828333291003248,"score_spread":0.3092976734262606,"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."}}