{"id":"W4405363817","doi":"10.1016/j.enbuild.2024.115177","title":"Explaining deep learning-based anomaly detection in energy consumption data by focusing on contextually relevant data","year":2024,"lang":"en","type":"article","venue":"Energy and Buildings","topic":"Big Data Technologies and Applications","field":"Decision Sciences","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada","keywords":"Anomaly detection; Anomaly (physics); Consumption (sociology); Energy consumption; Deep learning; Energy (signal processing); Computer science; Artificial intelligence; Data science; Machine learning; Engineering; Physics; Sociology; Statistics; Mathematics; Social science; Electrical engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.001433981,0.0001630579,0.0001962831,0.0003468417,0.0002761834,0.0005271299,0.001318648,0.0001662806,0.00005360673],"category_scores_gemma":[0.001080056,0.0001325456,0.00002136628,0.0006288717,0.0001435793,0.0007855203,0.0008071511,0.0002398973,0.00001339842],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003626765,"about_ca_system_score_gemma":0.00003196385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005390728,"about_ca_topic_score_gemma":0.001758507,"domain_scores_codex":[0.9976974,0.00009635279,0.0004256506,0.001130409,0.0004053921,0.0002448019],"domain_scores_gemma":[0.9973396,0.001131026,0.0001274072,0.001306394,0.00003736425,0.00005821449],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003513997,0.00003073859,0.0003162321,0.00000503215,0.00001005319,0.00001143773,0.00002739725,0.0004777839,0.009089825,0.01564286,0.005443584,0.9689099],"study_design_scores_gemma":[0.0002067795,0.00007475985,0.0001893387,0.00008395495,0.00001032872,0.00001001957,0.0001967358,0.3314343,0.006706523,0.001605241,0.6593059,0.0001760526],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2835527,0.006733546,0.705047,0.003042076,0.0003421407,0.00009366301,0.0003366715,0.0004829443,0.0003692117],"genre_scores_gemma":[0.9969937,0.0008491128,0.001160309,0.0002614701,0.00005727506,0.00001533097,0.0004973434,0.00001609036,0.0001493505],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9687338,"threshold_uncertainty_score":0.5405052,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1674095964407354,"score_gpt":0.3594454531349252,"score_spread":0.1920358566941898,"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."}}