{"paper":{"title":"QSMOTE-PGM/kPGM: QSMOTE Based PGM and kPGM for Imbalanced Dataset Classification","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"Quantum-inspired oversampling paired with pretty good measurement classifiers raises recall on imbalanced churn data over random forest.","cross_cats":["quant-ph"],"primary_cat":"cs.LG","authors_text":"Bikash K. Behera, Giuseppe Sergioli, Roberto Giuntini","submitted_at":"2025-12-18T07:36:26Z","abstract_excerpt":"Quantum-inspired machine learning (QiML) employs mathematical principles from quantum theory, such as Hilbert-space representations and quantum state discrimination, to enhance classical learning algorithms. In this work, we investigate the integration of Quantum Synthetic Minority Oversampling Technique (QSMOTE) variants with two quantum-inspired classifiers: the Pretty Good Measurement (PGM) classifier and the kernelized Pretty Good Measurement (KPGM) classifier. We propose and analyze three QSMOTE variants, namely KNN-based, Fidelity-based, and Margin-based QSMOTE, designed to improve minor"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental evaluations on the Telco Customer Churn dataset demonstrate that the proposed quantum-inspired approaches consistently outperform a classical Random Forest baseline, particularly in terms of recall and balanced F1-score. Among all configurations, PGM with stereo encoding and n_copies=2 achieves the best performance with an accuracy of 0.8512 and an F1-score of 0.8234.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the quantum-inspired similarity measures underlying the three QSMOTE variants (KNN-based, Fidelity-based, Margin-based) genuinely improve minority-class representation in a way that generalizes beyond the specific encodings and dataset tested, without post-hoc tuning or encoding-specific artifacts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"QSMOTE variants with PGM and KPGM classifiers outperform Random Forest on imbalanced Telco churn data, reaching 0.8512 accuracy and 0.8234 F1 using stereo encoding with two quantum copies.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Quantum-inspired oversampling paired with pretty good measurement classifiers raises recall on imbalanced churn data over random forest.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"85795af89224a9f054aacb805c4f8deb71c5c24b2698c8c039c063cf1f624979"},"source":{"id":"2512.16960","kind":"arxiv","version":2},"verdict":{"id":"984aae07-08f4-44e8-ade2-37c8c35047e3","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T21:54:20.445814Z","strongest_claim":"Experimental evaluations on the Telco Customer Churn dataset demonstrate that the proposed quantum-inspired approaches consistently outperform a classical Random Forest baseline, particularly in terms of recall and balanced F1-score. Among all configurations, PGM with stereo encoding and n_copies=2 achieves the best performance with an accuracy of 0.8512 and an F1-score of 0.8234.","one_line_summary":"QSMOTE variants with PGM and KPGM classifiers outperform Random Forest on imbalanced Telco churn data, reaching 0.8512 accuracy and 0.8234 F1 using stereo encoding with two quantum copies.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the quantum-inspired similarity measures underlying the three QSMOTE variants (KNN-based, Fidelity-based, Margin-based) genuinely improve minority-class representation in a way that generalizes beyond the specific encodings and dataset tested, without post-hoc tuning or encoding-specific artifacts.","pith_extraction_headline":"Quantum-inspired oversampling paired with pretty good measurement classifiers raises recall on imbalanced churn data over random forest."},"references":{"count":19,"sample":[{"doi":"10.1080/00107514","year":2015,"title":"An introduction to quantum machine learning,","work_id":"6fe14b8f-3479-437d-9760-4ffa5c2f4daa","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and S. Lloyd, “Quantum machine learning,”Nature, vol. 549, no. 7671, pp. 195–202, 2017","work_id":"5262c1da-7bf0-43b3-b584-4ed2247721c5","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1994,"title":"P. Hausladen and W. K. Wootters, “Pretty good measurement,”Journal of Modern Optics, vol. 41, no. 12, pp. 2385–2390, 1994","work_id":"6d1f758d-4189-4837-91de-63e1ac0ef9f7","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"A new quantum approach to binary classification,","work_id":"69f9d47b-0a2f-4b57-ad61-03b13b7f336d","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"10.1016/j.asoc.2022.109956","year":2023,"title":"Quantum-inspired algorithm for direct multi-class classification,","work_id":"d4c447ac-a9e3-42e5-be71-53525a65970a","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":19,"snapshot_sha256":"3222aa7ad660c2f1c1f95f7ac566075798829a6186ac78ec8a6ddf6f498c6521","internal_anchors":1},"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"}