{"paper":{"title":"Comparative Evaluation of Machine Learning Approaches for Minority-Class Financial Distress Prediction Under Class Imbalance Constraints","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Gradient-boosting models achieve higher sensitivity to rare financial distress cases than statistical baselines under severe class imbalance.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Karan Sehgal, Khawar Naveed Bhatti","submitted_at":"2026-05-13T19:44:25Z","abstract_excerpt":"Financial distress prediction remains a significant challenge in enterprise risk analysis due to the highly imbalanced nature of real-world financial datasets, where bankrupt or distressed firms typically constitute only a small minority of observations.\n  This paper presents a comparative evaluation of classical statistical methods, ensemble learning approaches, and exploratory neural models for minority-class financial distress prediction under class imbalance constraints.\n  The study incorporates structured preprocessing, imbalance mitigation using the Synthetic Minority Oversampling Techni"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental evaluation demonstrates that gradient-boosting approaches achieved improved minority-class sensitivity relative to baseline statistical classifiers under severe imbalance conditions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the chosen financial datasets and SMOTE-generated samples are representative of real-world distress distributions and that performance gains are not artifacts of the synthetic oversampling process.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Gradient boosting models with SMOTE oversampling show better minority-class sensitivity than statistical baselines for financial distress prediction under severe imbalance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Gradient-boosting models achieve higher sensitivity to rare financial distress cases than statistical baselines under severe class imbalance.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d4053353b22cd6e84041e26a1f006d4665a7a1823db2a7972b77b8c851c4d0a7"},"source":{"id":"2605.14067","kind":"arxiv","version":1},"verdict":{"id":"1561456c-cdea-43e9-abc3-48cac157a039","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T05:21:31.539646Z","strongest_claim":"Experimental evaluation demonstrates that gradient-boosting approaches achieved improved minority-class sensitivity relative to baseline statistical classifiers under severe imbalance conditions.","one_line_summary":"Gradient boosting models with SMOTE oversampling show better minority-class sensitivity than statistical baselines for financial distress prediction under severe imbalance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the chosen financial datasets and SMOTE-generated samples are representative of real-world distress distributions and that performance gains are not artifacts of the synthetic oversampling process.","pith_extraction_headline":"Gradient-boosting models achieve higher sensitivity to rare financial distress cases than statistical baselines under severe class imbalance."},"references":{"count":8,"sample":[{"doi":"","year":null,"title":"The Journal of Finance , volume=","work_id":"c1a49aac-6cca-47ce-9880-2567cd281c3b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Journal of Accounting Research , volume=","work_id":"2f73e04a-b0c7-46c1-a9e1-a9732cbd4653","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Journal of Accounting Research , volume=","work_id":"40cc2c8b-15c8-4638-82b5-2c14f591adf6","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , pages=","work_id":"34c7400d-a675-4efb-9cb6-8fc36b741507","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Advances in Neural Information Processing Systems , volume=","work_id":"5cc2f44f-1087-4dab-8a8a-f14a24d55eae","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":8,"snapshot_sha256":"a9625c6200c87e738c51f04e9d2def09006314e494d2adc059a4322e550325f9","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}