{"id":"W4304757181","doi":"10.1093/bioinformatics/btac649","title":"METAbolomics data Balancing with Over-sampling Algorithms (META-BOA): an online resource for addressing class imbalance","year":2022,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada; University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Overfitting; Sampling (signal processing); Class (philosophy); Data mining; Machine learning; Sample (material); Visualization; Random forest; Resource (disambiguation); Artificial intelligence; Algorithm; Artificial neural network","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007168158,0.002697004,0.002076896,0.004357685,0.001382647,0.003466852,0.004394562,0.002146498,0.06572302],"category_scores_gemma":[0.02396043,0.001585709,0.003008183,0.003723949,0.0005981429,0.003808237,0.00528297,0.002995172,0.04090405],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009737501,"about_ca_system_score_gemma":0.00266424,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002884648,"about_ca_topic_score_gemma":0.005453579,"domain_scores_codex":[0.9971943,0.0005250673,0.0002527017,0.0007130894,0.001139655,0.0001751817],"domain_scores_gemma":[0.9899216,0.005072106,0.0007400602,0.002121118,0.001596109,0.0005491448],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002063764,0.0004303207,0.00885072,0.00260855,0.0006620751,0.0004332018,0.0004438718,0.00768952,0.01326726,0.006616671,0.7250973,0.2318368],"study_design_scores_gemma":[0.001499514,0.0004208888,0.01841026,0.001273556,0.0003948296,0.001036697,0.0002837737,0.2599527,0.06030307,0.07375675,0.5819941,0.0006740035],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"software","genre_gemma":"methods","genre_scores_codex":[0.006084862,0.001323235,0.287792,0.001501445,0.0005911322,0.0006767624,0.09368397,0.6004459,0.007900686],"genre_scores_gemma":[0.057655,0.001390617,0.6352488,0.00299027,0.0004491551,0.003736451,0.1823909,0.1059793,0.01015951],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.06572302,"threshold_uncertainty_score":0.2198654,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1940735828464191,"score_gpt":0.3426043390083074,"score_spread":0.1485307561618883,"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."}}