{"id":"W2533147238","doi":"10.1016/j.dib.2016.09.032","title":"Infodemiological data of high-school drop-out related web searches in Canada correlating with real-world statistical data in the period 2004–2012","year":2016,"lang":"en","type":"article","venue":"Data in Brief","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Drop (telecommunication); Percentage point; Statistics; Lag; Demography; Drop out; Mathematics; Geography; Demographic economics; Computer science; Economics; Sociology","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.003188392,0.0002450518,0.0006960598,0.0001461659,0.00003660718,0.00002182886,0.003123183,0.00009136608,0.0006467561],"category_scores_gemma":[0.009080376,0.0001413147,0.000008959893,0.0006250214,0.0004115132,0.0008546962,0.002560963,0.0007501327,0.00003366588],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003899185,"about_ca_system_score_gemma":0.003431779,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.7123034,"about_ca_topic_score_gemma":0.9714566,"domain_scores_codex":[0.9959269,0.0007063984,0.0009662402,0.001048241,0.0007582984,0.000593936],"domain_scores_gemma":[0.9905122,0.002165646,0.0002255235,0.006807772,0.00005380133,0.0002350409],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0004940337,0.0001479037,0.8980165,0.00007499505,0.0000398803,0.0008936145,0.00005370049,0.00001062811,0.0000443774,0.0004385538,0.09567225,0.004113547],"study_design_scores_gemma":[0.002953846,0.00007008808,0.9646696,0.0005569337,0.00004074804,0.00003984269,0.0002103325,0.00838043,0.000001490484,0.00006914091,0.02278891,0.0002186306],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7270571,0.0007703308,0.0008558148,0.007670206,0.000253883,0.001851364,0.258929,0.00007883186,0.002533537],"genre_scores_gemma":[0.9273915,0.0002135634,0.002366777,0.0004008094,0.00005292219,0.00001384701,0.06942189,0.00002695385,0.0001117558],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2591532,"threshold_uncertainty_score":0.9992666,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07115387420654301,"score_gpt":0.3269425574381863,"score_spread":0.2557886832316433,"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."}}