{"id":"W4416075025","doi":"10.3390/jrfm18110630","title":"Optimizing the Collection Process in Credit Risk Management: A Comparison of Machine Learning Techniques for Predicting Payment Probability at Different Stages of Arrears","year":2025,"lang":"en","type":"article","venue":"Journal of risk and financial management","topic":"Financial Distress and Bankruptcy Prediction","field":"Business, Management and Accounting","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Arrears; Artificial neural network; Logistic regression; Metric (unit); Credit risk; Process (computing); Reliability (semiconductor); Feature selection","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":[0.001061044,0.0001875719,0.0004702204,0.000490832,0.0003525852,0.00005843874,0.0002059242,0.0000692848,0.000004392315],"category_scores_gemma":[0.0002090424,0.0001366781,0.0001456408,0.0005494837,0.00008161127,0.0002615532,0.0002261263,0.0002835507,9.76861e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001000604,"about_ca_system_score_gemma":0.00001078474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002406809,"about_ca_topic_score_gemma":0.0003506677,"domain_scores_codex":[0.9982637,0.00004598544,0.0009506401,0.0002181133,0.0003141654,0.0002073969],"domain_scores_gemma":[0.9981081,0.00009297292,0.001453187,0.0001307748,0.0002050038,0.00000995994],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00121387,0.0004891352,0.8331791,0.002538358,0.00008714203,0.000002902113,0.0006681732,0.003663169,0.00002416502,0.005018707,0.0002427671,0.1528725],"study_design_scores_gemma":[0.00312715,0.0003660572,0.9420922,0.001945355,0.000999626,8.469486e-7,0.002581774,0.01962777,0.00102291,0.01468345,0.01328287,0.0002699577],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9593458,0.0008979186,0.03725954,0.00008472828,0.0003695023,0.001176866,0.00002078966,0.00002433472,0.000820497],"genre_scores_gemma":[0.9968733,0.001546602,0.001268648,0.00001663205,0.0001450314,0.00007838951,0.000007832002,0.00001086555,0.00005275631],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1526025,"threshold_uncertainty_score":0.5573571,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00927671189483235,"score_gpt":0.2402195637802324,"score_spread":0.2309428518854001,"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."}}