{"id":"W4410734606","doi":"10.1002/for.3284","title":"Default Prediction Framework With Optimal Feature Set and Matching Ratio","year":2025,"lang":"en","type":"article","venue":"Journal of Forecasting","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"National Natural Science Foundation of China","keywords":"Feature (linguistics); Matching (statistics); Computer science; Set (abstract data type); Pattern recognition (psychology); Artificial intelligence; Econometrics; Mathematics; Statistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00356605,0.000984504,0.001637817,0.001862037,0.0005796344,0.001257636,0.002351975,0.001177797,0.001735286],"category_scores_gemma":[0.007771081,0.0003668866,0.0007233505,0.001281074,0.0005138455,0.002412776,0.001060412,0.001429215,0.0004878232],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001144369,"about_ca_system_score_gemma":0.001215891,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005817152,"about_ca_topic_score_gemma":0.003872156,"domain_scores_codex":[0.9987236,0.0003800458,0.00007643586,0.0003593345,0.0002852458,0.0001752762],"domain_scores_gemma":[0.9980034,0.0009525403,0.0002307597,0.0002216685,0.0004874293,0.0001041814],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008444559,0.0006297724,0.0200484,0.00009884895,0.0001884572,0.00044661,0.0001328103,0.6103042,0.003270279,0.01115484,0.01043347,0.3424478],"study_design_scores_gemma":[0.000009860329,0.00001778199,0.0004116868,0.00000364994,0.000008423994,0.00001282254,0.000005213003,0.9953929,0.0002531106,0.003770441,0.0001099319,0.000004122716],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1935092,0.0007885661,0.7988015,0.00110436,0.00009481379,0.0001496786,0.0007657217,0.002194614,0.002591537],"genre_scores_gemma":[0.9434149,0.0001031535,0.05408428,0.0001736861,0.00008212334,0.00011958,0.0008383405,0.00004963913,0.001134411],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005817152,"threshold_uncertainty_score":0.01885933,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01887874567049567,"score_gpt":0.2666547613807316,"score_spread":0.2477760157102359,"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."}}