{"id":"W4296640192","doi":"10.1016/j.buildenv.2022.109620","title":"Unsupervised outlier detection using neural network-based mixtures of probabilistic principal component analyzers for building chiller fault diagnosis","year":2022,"lang":"en","type":"article","venue":"Building and Environment","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Outlier; Anomaly detection; Principal component analysis; Computer science; Probabilistic logic; Fault detection and isolation; Artificial neural network; Data mining; Chiller boiler system; Artificial intelligence; Fault (geology); Pattern recognition (psychology); Engineering; Water chiller","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":[0.0002775822,0.0001363554,0.0001660697,0.00008570854,0.0005939524,0.00003646918,0.0002380717,0.00003501019,0.00001568291],"category_scores_gemma":[0.000007930009,0.000138574,0.00009724093,0.0001644201,0.00005333476,0.00006339191,0.0002345119,0.000111324,1.346357e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000145616,"about_ca_system_score_gemma":0.00001093908,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005022591,"about_ca_topic_score_gemma":0.000001139953,"domain_scores_codex":[0.9988585,0.00006120998,0.0002599267,0.0003894042,0.0002054177,0.0002255956],"domain_scores_gemma":[0.9994132,0.0000857506,0.0001467696,0.0002789291,0.00001009065,0.00006528509],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001564704,0.00009979683,0.001672782,0.00002154144,0.00001820791,6.442689e-7,0.00006052909,0.9596949,0.01728968,0.0008778722,0.00001700641,0.02023138],"study_design_scores_gemma":[0.0003250465,0.0001963612,0.003744684,0.00001112036,0.00003539542,0.000006432934,0.00001538396,0.9765289,0.01379254,0.0007033132,0.004457343,0.0001835305],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5282786,0.00008687931,0.4710604,0.0001329794,0.00004976516,0.0003344583,0.000006271714,0.0000480926,0.000002597173],"genre_scores_gemma":[0.902352,0.00001505182,0.0969455,0.00008352425,0.00003855607,0.0005447079,0.000003037668,0.00001234853,0.000005271529],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3741149,"threshold_uncertainty_score":0.5650882,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02076974901253803,"score_gpt":0.2429634756932868,"score_spread":0.2221937266807488,"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."}}