{"id":"W3093643609","doi":"10.2196/21209","title":"Understanding User Behavior Through the Use of Unsupervised Anomaly Detection: Proof of Concept Using Internet of Things Smart Home Thermostat Data for Improving Public Health Surveillance","year":2020,"lang":"en","type":"article","venue":"JMIR mhealth and uhealth","topic":"Context-Aware Activity Recognition Systems","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; University of Toronto; University of Waterloo","funders":"","keywords":"Computer science; Internet privacy; The Internet; Wearable computer; Event (particle physics); Anomaly detection; Public health; Ubiquitous computing; Tracking (education); Computer security; Behavioral pattern; Public health surveillance; Activity recognition; Mental health; Human–computer interaction; World Wide Web; Psychology; Data mining; Medicine; Artificial intelligence; Psychiatry","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001336859,0.000207879,0.0006718467,0.0000953619,0.0002306932,0.000104439,0.0007684887,0.00008828944,0.000003324046],"category_scores_gemma":[0.0001623125,0.0001685789,0.00007320328,0.0006013088,0.0001912701,0.001704733,0.0003990033,0.0002282567,1.369942e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001784778,"about_ca_system_score_gemma":0.001058273,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003802054,"about_ca_topic_score_gemma":0.0004400498,"domain_scores_codex":[0.9969536,0.0005261624,0.001038146,0.0006121892,0.000378774,0.0004911452],"domain_scores_gemma":[0.9966296,0.0006768494,0.001355968,0.0008001179,0.0002711697,0.0002663202],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001087735,0.001403023,0.1118127,0.02918991,0.0003468207,0.000007432846,0.1776192,0.00008179858,0.004111399,0.009214608,0.001495744,0.6636297],"study_design_scores_gemma":[0.009845105,0.008038474,0.04504224,0.001046768,0.0001628673,0.0001741108,0.01535718,0.9056066,0.004202143,0.0004955527,0.00851221,0.00151677],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2893465,0.0004687273,0.7044608,0.003582862,0.0002178447,0.001665164,0.0002078714,0.00004681146,0.000003385481],"genre_scores_gemma":[0.9920149,0.0000323196,0.006500205,0.001302056,0.00005270095,0.00004049151,0.0000297622,0.00002345147,0.000004140696],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9055248,"threshold_uncertainty_score":0.6874447,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4900413572962249,"score_gpt":0.3793201948349918,"score_spread":0.1107211624612331,"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."}}