{"id":"W2906950933","doi":"10.1016/j.eswa.2019.01.011","title":"Deep understanding in industrial processes by complementing human expertise with interpretable patterns of machine learning","year":2019,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Fault Detection and Control Systems","field":"Engineering","cited_by":41,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal; Natural Resources Canada","funders":"Natural Resources Canada; Institut de Valorisation des Données","keywords":"Computer science; Fault tree analysis; Process (computing); Domain knowledge; Artificial intelligence; Data mining; Set (abstract data type); Machine learning; Domain (mathematical analysis); Fault (geology); Tree (set theory); Decision tree; Knowledge extraction; Engineering; Reliability 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":[0.0001161206,0.0001741428,0.0003113469,0.0001334878,0.00009338748,0.0000607302,0.0001395797,0.00006521042,0.00003838442],"category_scores_gemma":[0.000004992005,0.0001438923,0.00001929916,0.0003325375,0.00001628339,0.0001407652,0.00001472662,0.0001822491,0.000007499051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002033933,"about_ca_system_score_gemma":0.00001765473,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001223539,"about_ca_topic_score_gemma":0.0007273572,"domain_scores_codex":[0.998901,0.00004071217,0.0004016192,0.0002235137,0.0001999154,0.0002332296],"domain_scores_gemma":[0.9995028,0.00005651266,0.0001314855,0.0002139197,0.00003980201,0.00005549542],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00026836,0.0003895705,0.5288599,0.003206884,0.0007272373,0.000006366731,0.02088283,0.2694381,0.1703418,0.003336995,0.0006792902,0.00186266],"study_design_scores_gemma":[0.008325249,0.0006904781,0.0002351484,0.003791086,0.00004442709,0.00006888339,0.06284685,0.8655143,0.007756106,0.00001017453,0.04927893,0.001438346],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2484355,0.002660675,0.7398005,0.00003341251,0.0001793771,0.003358832,0.00003228384,0.0005293454,0.004970034],"genre_scores_gemma":[0.998395,0.00001069466,0.00003871058,0.000005659321,0.00005622184,0.001213662,0.000036393,0.00004625692,0.000197391],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7499595,"threshold_uncertainty_score":0.5867757,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01908878235177313,"score_gpt":0.2344379540207849,"score_spread":0.2153491716690118,"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."}}