{"id":"W4389058774","doi":"10.3390/jrfm16120496","title":"Machine Learning for Enhanced Credit Risk Assessment: An Empirical Approach","year":2023,"lang":"en","type":"article","venue":"Journal of risk and financial management","topic":"Financial Distress and Bankruptcy Prediction","field":"Business, Management and Accounting","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Office of Energy Research and Development; Conselho Nacional de Desenvolvimento Científico e Tecnológico; Fundação de Amparo à Pesquisa do Estado de São Paulo","keywords":"Machine learning; Computer science; Artificial intelligence; Credit risk; Logistic regression; Financial institution; Exploratory data analysis; Key (lock); Finance; Data mining; Business","routes":{"ca_aff":false,"ca_fund":true,"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":[0.001177293,0.0001925933,0.0003268856,0.0004584194,0.0005393262,0.0002501773,0.0002003405,0.00008902979,0.00001261131],"category_scores_gemma":[0.000235264,0.0001628557,0.0001632397,0.0005586778,0.00004238649,0.0008826432,0.0001402373,0.0003686259,0.00001174322],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003048017,"about_ca_system_score_gemma":0.00001726822,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008081325,"about_ca_topic_score_gemma":0.00002477635,"domain_scores_codex":[0.9985671,0.00003106547,0.0004889203,0.0002635636,0.0003422769,0.0003070495],"domain_scores_gemma":[0.9989284,0.0000480964,0.0006813333,0.0001240655,0.000187535,0.00003056746],"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":[0.0006440943,0.0004721366,0.1187394,0.0005363298,0.00007913135,0.00004570851,0.0002893177,0.005093493,0.0000646008,0.01660736,0.009340654,0.8480878],"study_design_scores_gemma":[0.001999452,0.0002106407,0.586479,0.00006226563,0.0003566053,0.000002252623,0.0004360682,0.05152582,0.000005093641,0.01227388,0.3463875,0.0002613765],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5266798,0.0002567336,0.4658931,0.0001831105,0.00140966,0.00051888,0.00004744626,0.0001505544,0.004860768],"genre_scores_gemma":[0.9927702,0.00137855,0.002699318,0.0001428195,0.002682095,0.00003110831,0.00009534383,0.00002851512,0.0001720747],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8478264,"threshold_uncertainty_score":0.6641062,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01624442269174582,"score_gpt":0.261793363822958,"score_spread":0.2455489411312122,"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."}}