{"id":"W4286850364","doi":"10.1016/j.eswa.2022.118809","title":"On the combination of graph data for assessing thin-file borrowers' creditworthiness","year":2021,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Financial Distress and Bankruptcy Prediction","field":"Business, Management and Accounting","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"Western University","funders":"","keywords":"Computer science; Feature engineering; Graph; Machine learning; Feature (linguistics); Feature learning; Population; Representation (politics); Artificial neural network; Artificial intelligence; Data science; Data mining; Deep learning; Theoretical computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002258048,0.00009908029,0.0001208768,0.00008787278,0.0002689079,0.0001241181,0.0003924154,0.00005598946,0.0002985689],"category_scores_gemma":[0.0003581796,0.00008731174,0.00007048614,0.0006442663,0.00006087947,0.001233795,0.000213731,0.00008818258,0.0000128934],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001634784,"about_ca_system_score_gemma":0.00003514102,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000147051,"about_ca_topic_score_gemma":0.00009833049,"domain_scores_codex":[0.9993364,0.00001447562,0.0001186963,0.0003173098,0.00007801534,0.0001350953],"domain_scores_gemma":[0.9988175,0.0001717574,0.000187578,0.000507259,0.0003091663,0.000006736249],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004948781,0.0001776305,0.003049565,0.0000921797,0.0000255971,0.00001740666,0.000009824979,0.002192296,0.00008246623,0.9601709,0.03366695,0.0004657177],"study_design_scores_gemma":[0.002158822,0.00004128646,0.08287479,0.0003782116,0.000388334,0.000001261376,0.001607155,0.5996724,0.0002533749,0.2823043,0.029784,0.0005360265],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9138288,0.00003253895,0.05962619,0.0006416995,0.001141388,0.0003361835,0.0003562657,0.0001218833,0.02391507],"genre_scores_gemma":[0.9982613,0.000008069716,0.00002754833,0.0002671266,0.0001980471,9.500289e-7,0.0008995392,0.00001037817,0.0003270419],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6778666,"threshold_uncertainty_score":0.3560469,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07564071058433029,"score_gpt":0.193105622686374,"score_spread":0.1174649121020437,"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."}}