{"id":"W2787468747","doi":"10.3390/jrfm11010012","title":"Ensemble Learning or Deep Learning? Application to Default Risk Analysis","year":2018,"lang":"en","type":"article","venue":"Journal of risk and financial management","topic":"Financial Distress and Bankruptcy Prediction","field":"Business, Management and Accounting","cited_by":118,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Boosting (machine learning); Credit risk; Artificial neural network; Artificial intelligence; Gradient boosting; Computer science; Ensemble learning; Random forest; Machine learning; Payment; Dropout (neural networks); Default; Econometrics; Actuarial science; Business; Finance; Economics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007699657,0.0001747759,0.0003292354,0.0007439766,0.0005894709,0.0002101478,0.0001841503,0.00007532753,0.00005750476],"category_scores_gemma":[0.0003885067,0.0001405463,0.0001633327,0.001454545,0.00004643816,0.0004465322,0.000157584,0.0003137258,0.0000888312],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000374033,"about_ca_system_score_gemma":0.000009269304,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005214851,"about_ca_topic_score_gemma":0.00092077,"domain_scores_codex":[0.9986442,0.00003028747,0.0004811295,0.0002555848,0.0003325024,0.0002563328],"domain_scores_gemma":[0.998693,0.00003267026,0.0007802329,0.0001339744,0.0003258504,0.00003421157],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0005375892,0.00009062387,0.1706934,0.00005682718,0.0001179706,0.00002104326,0.0002416589,0.004455832,0.00002316504,0.003123115,0.0008079027,0.8198309],"study_design_scores_gemma":[0.0005665303,0.0001915587,0.5866599,0.0000366749,0.001302783,0.000002011108,0.000367768,0.009298043,0.000009986391,0.001415618,0.3999534,0.0001957126],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4746322,0.0001170895,0.5222781,0.00005763209,0.0003008564,0.0001808711,0.000001930318,0.00003952226,0.002391806],"genre_scores_gemma":[0.9957783,0.0006302648,0.001160094,0.0002016998,0.001937146,0.00001225367,0.000007066665,0.00001613423,0.0002570742],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8196352,"threshold_uncertainty_score":0.5731311,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004770270859417123,"score_gpt":0.2107345050049583,"score_spread":0.2059642341455412,"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."}}