{"id":"W4401326743","doi":"10.1109/access.2024.3439248","title":"Enhancing Transformer Health Index Prediction Using Dissolved Gas Analysis Data Through Integration of LightGBM and Robust EM Algorithms","year":2024,"lang":"en","type":"article","venue":"IEEE Access","topic":"Power Transformer Diagnostics and Insulation","field":"Engineering","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"Natural Sciences and Engineering Research Council of Canada; Saskatchewan Research Council","keywords":"Missing data; Computer science; Imputation (statistics); Data mining; Boosting (machine learning); Dissolved gas analysis; Computation; Expectation–maximization algorithm; Maximization; Data modeling; Transformer; Algorithm; Artificial intelligence; Machine learning; Statistics; Maximum likelihood; Engineering; Mathematics; Mathematical optimization; Transformer oil","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":[0.0002071353,0.0001383807,0.0002331711,0.0002319409,0.00006314499,0.0001695209,0.0001650235,0.00007353467,0.00001446866],"category_scores_gemma":[0.000007603254,0.0001278295,0.00004974383,0.00078443,0.00001892473,0.001736276,0.00001046414,0.0001391051,5.750729e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006902523,"about_ca_system_score_gemma":0.00003755317,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006237865,"about_ca_topic_score_gemma":0.001232938,"domain_scores_codex":[0.9989782,0.00001518178,0.0004057063,0.0002466273,0.0001780388,0.0001762278],"domain_scores_gemma":[0.9995983,0.0000513416,0.00003536764,0.0002284745,0.00003725593,0.00004928319],"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":[0.00003218869,0.0001175105,0.006700446,0.001993128,0.002254145,0.000006883623,0.01479541,0.7850906,0.06019844,0.0002750746,0.0004389648,0.1280973],"study_design_scores_gemma":[0.0001182517,0.00002287922,0.007446139,0.0002569807,0.0002936553,0.000001843086,0.0001100393,0.9664271,0.02497406,0.0001281613,0.0001046876,0.0001161932],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2897187,0.001070936,0.7081984,0.00003560957,0.0004659141,0.0001368101,0.0002168744,0.00009086022,0.00006591383],"genre_scores_gemma":[0.9957229,0.001782487,0.001975404,0.00001467462,0.0001070963,0.000006002018,0.0003648705,0.00002418728,0.000002336925],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.706223,"threshold_uncertainty_score":0.5212734,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06293758002059877,"score_gpt":0.326979939128666,"score_spread":0.2640423591080672,"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."}}