{"id":"W4283741364","doi":"10.1002/for.2885","title":"Predicting earnings management through machine learning ensemble classifiers","year":2022,"lang":"en","type":"article","venue":"Journal of Forecasting","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"Saint Mary's University; Concordia University","funders":"Social Sciences and Humanities Research Council of Canada; New Brunswick Innovation Foundation","keywords":"Artificial intelligence; Computer science; Ensemble forecasting; Support vector machine; Ensemble learning; Machine learning; Principal component analysis; Random subspace method; Context (archaeology); Feature selection; Pattern recognition (psychology); Classifier (UML)","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":[0.003076887,0.000865639,0.001190266,0.002737385,0.0004441731,0.00103773,0.0008700532,0.0008018505,0.0008313642],"category_scores_gemma":[0.00701612,0.0002037574,0.0007669674,0.001442607,0.0001718551,0.00142072,0.0006048861,0.001135361,0.0004267636],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005734395,"about_ca_system_score_gemma":0.0005390772,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0053101,"about_ca_topic_score_gemma":0.004419303,"domain_scores_codex":[0.9990243,0.0002552443,0.00007914033,0.0001763243,0.0003395963,0.0001252551],"domain_scores_gemma":[0.9955164,0.002220057,0.0005150848,0.0003152302,0.001284284,0.0001490356],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003029534,0.0004401813,0.1247214,0.00006063555,0.0005370122,0.0001780282,0.00009729674,0.3963313,0.002292506,0.001551672,0.006281935,0.467205],"study_design_scores_gemma":[0.000004492481,0.00004772734,0.005701232,0.00001431398,0.00003855171,0.00001718445,0.00002369308,0.9921805,0.0007246886,0.0008488636,0.0003907393,0.000008010946],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6556103,0.002318952,0.3332534,0.000880305,0.0003991589,0.0001245054,0.0008777419,0.001197731,0.005337938],"genre_scores_gemma":[0.9680618,0.0002527087,0.02994423,0.00007440955,0.0001285569,0.00003624561,0.0005868027,0.00001605032,0.0008991659],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0053101,"threshold_uncertainty_score":0.01627231,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04131222352818766,"score_gpt":0.2561634504400872,"score_spread":0.2148512269118995,"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."}}