{"id":"W4289544976","doi":"10.1186/s43065-022-00055-4","title":"Multiclass anomaly detection in imbalanced structural health monitoring data using convolutional neural network","year":2022,"lang":"en","type":"article","venue":"Journal of Infrastructure Preservation and Resilience","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"University of Illinois at Urbana-Champaign; Ministry of Colleges and Universities; Western University; Harbin Institute of Technology","keywords":"Anomaly detection; Computer science; Structural health monitoring; Convolutional neural network; Hyperparameter; Data mining; Anomaly (physics); Big data; Machine learning; Artificial intelligence; 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":[0.0004729345,0.0001568078,0.0002533914,0.0001719561,0.0003119503,0.00005191977,0.0004081075,0.00005046887,0.00001305524],"category_scores_gemma":[0.00007331205,0.0001433241,0.00003200925,0.0003931484,0.00004991585,0.0009037945,0.000222705,0.0006895328,4.901866e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003360307,"about_ca_system_score_gemma":0.00009488573,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005472791,"about_ca_topic_score_gemma":0.00002229916,"domain_scores_codex":[0.9984196,0.00009826099,0.0005813952,0.000181117,0.0003872061,0.0003324164],"domain_scores_gemma":[0.9992526,0.00005256262,0.0002981039,0.000213094,0.00009262635,0.00009106663],"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":[0.00005608812,0.000002153929,0.1412192,0.00003066451,0.00001307891,0.000008116519,0.0002741498,0.8397,0.01041921,0.00002013775,0.00008730099,0.00816993],"study_design_scores_gemma":[0.0004532894,0.00005612501,0.470636,0.0000405092,0.000005321584,0.0002190633,0.0003688779,0.5270345,0.0003691654,0.0003832561,0.0003276373,0.0001062478],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9905192,0.00141162,0.005350417,0.00008114638,0.002456131,0.0001310697,0.00001404181,0.00002495634,0.00001143219],"genre_scores_gemma":[0.9909577,0.00008771859,0.008218099,0.00004500505,0.0006638003,0.000002420317,0.000005831175,0.00001586953,0.00000360741],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3294169,"threshold_uncertainty_score":0.5844588,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01849641497267391,"score_gpt":0.2715726988179669,"score_spread":0.253076283845293,"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."}}