{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T09:55:16Z","timestamp":1781690116880,"version":"3.54.5"},"reference-count":27,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T00:00:00Z","timestamp":1778803200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T00:00:00Z","timestamp":1778803200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Expert Systems"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>Modern e\u2010commerce platforms face a critical challenge: delivering accurate recommendations under extreme user\u2013item interaction sparsity, where textual context remains systematically underutilised. Existing collaborative filtering methods degrade sharply in sparse settings, while semantic approaches fail to capture collaborative patterns effectively. We propose CF\u2010SBERTHet, a unified recommendation framework that integrates collaborative filtering signals with rich semantic information extracted from user reviews through heterogeneous graph neural networks. Our approach generates enriched item embeddings by aligning representations from a pre\u2010trained collaborative filtering model with semantic features extracted via Sentence\u2010BERT, and then models complex multi\u2010relational interactions through type\u2010specific message passing over seven distinct edge types, encompassing ratings, reviews, purchases, and contextual item similarities. Extensive experiments on four Amazon 2023 datasets (Fashion, Beauty, Musical Instruments and Movies and TV) demonstrate that CF\u2010SBERTHet consistently outperforms competitive baselines across all evaluation settings in terms of both Root Mean Square Error (RMSE) and Mean Absolute Error (MAE), achieving robust performance even at interaction sparsities exceeding 99%. Comprehensive ablation studies confirm that each component, namely the collaborative and textual knowledge\u2010enriched embeddings, the heterogeneous graph architecture, and the review\u2010based edge weights, contributes critically to overall performance. These results establish CF\u2010SBERTHet as an effective and broadly applicable solution for context\u2010aware recommendation in sparse real\u2010world e\u2010commerce environments.<\/jats:p>","DOI":"10.1111\/exsy.70297","type":"journal-article","created":{"date-parts":[[2026,5,15]],"date-time":"2026-05-15T11:46:37Z","timestamp":1778845597000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["<scp>CF<\/scp>\n                    \u2010\n                    <scp>SBERTHet<\/scp>\n                    : Collaborative and Textual Knowledge Enhanced Semantic Graphs for Sparse Recommendations"],"prefix":"10.1111","volume":"43","author":[{"given":"He","family":"Ma","sequence":"first","affiliation":[{"name":"School of Computer Science The University of Sydney  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5071-7515","authenticated-orcid":false,"given":"Maryam Khanian","family":"Najafabadi","sequence":"additional","affiliation":[{"name":"Computer Science and Data Science Australian Catholic University  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qiyang","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Computer Science The University of Sydney  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"WeiPing","family":"Kong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering The University of New South Wales  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huajun","family":"Jie","sequence":"additional","affiliation":[{"name":"School of Computer Science The University of Sydney  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fanye","family":"Sun","sequence":"additional","affiliation":[{"name":"College of Engineering Northeastern University  Boston Massachusetts USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoyue","family":"Bai","sequence":"additional","affiliation":[{"name":"School of Computer Science The University of Sydney  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mohammadhossein","family":"Ahmadi","sequence":"additional","affiliation":[{"name":"School of Computing, Faculty of Science and Engineering Macquarie University  Sydney New South Wales Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,5,15]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.125667"},{"key":"e_1_2_10_3_1","doi-asserted-by":"crossref","unstructured":"Chen Y. J.Li andC.Xiong.2022.\u201cELECRec: Training Sequential Recommenders as Discriminators.\u201dInProceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval 2550\u20132554.","DOI":"10.1145\/3477495.3531894"},{"key":"e_1_2_10_4_1","unstructured":"Chen Y. X.Sun andZ.Liu.2024.\u201cLLM\u2010Guided Multi\u2010View Hypergraph Learning for Human\u2010Centric Explainable Recommendation.\u201darXiv Preprint arXiv:2401.08217."},{"key":"e_1_2_10_5_1","doi-asserted-by":"crossref","unstructured":"Cheng H. T. L.Koc J.Harmsen et\u00a0al.2016.\u201cWide & Deep Learning for Recommender Systems.\u201dInProceedings of the 1st Workshop on Deep Learning for Recommender Systems 7\u201310.","DOI":"10.1145\/2988450.2988454"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.330194"},{"key":"e_1_2_10_7_1","doi-asserted-by":"crossref","unstructured":"He X. K.Deng X.Wang Y.Li Y.Zhang andM.Wang.2020.\u201cLightgcn: Simplifying and Powering Graph Convolution Network for Recommendation.\u201dInProceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval 639\u2013648.","DOI":"10.1145\/3397271.3401063"},{"key":"e_1_2_10_8_1","doi-asserted-by":"crossref","unstructured":"Kang W. C. andJ.McAuley.2018.\u201cSelf\u2010Attentive Sequential Recommendation.\u201dIn2018 IEEE International Conference on Data Mining (ICDM). IEEE 197\u2013206.","DOI":"10.1109\/ICDM.2018.00035"},{"key":"e_1_2_10_9_1","doi-asserted-by":"crossref","unstructured":"Kim S. H.Kang S.Choi D.Kim M.Yang andC.Park.2024.\u201cLarge Language Models Meet Collaborative Filtering: An Efficient All\u2010Round LLM\u2010Based Recommender System.\u201dInProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining 1395\u20131406.","DOI":"10.1145\/3637528.3671931"},{"key":"e_1_2_10_10_1","unstructured":"Li J. K.Wang andY.Zhou.2024.\u201cIntegrating Large Language Models With Graphical Session\u2010Based Recommendation.\u201darXiv Preprint arXiv:2402.16539."},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.70041"},{"key":"e_1_2_10_12_1","unstructured":"Liu J. C.Liu P.Zhou R.Lv K.Zhou andY.Zhang.2023.\u201cIs Chatgpt a Good Recommender? A Preliminary Study.\u201darXiv Preprint arXiv: 2304.10149."},{"key":"e_1_2_10_13_1","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00619"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3177611"},{"key":"e_1_2_10_15_1","unstructured":"Ma J. C.Zhou P.Cui H.Yang andW.Zhu.2019.\u201cLearning Disentangled Representations for Recommendation.\u201dAdvances in Neural Information Processing Systems 32."},{"key":"e_1_2_10_16_1","doi-asserted-by":"crossref","unstructured":"McAuley J. andJ.Leskovec.2013.\u201cHidden Factors and Hidden Topics: Understanding Rating Dimensions With Review Text.\u201dInProceedings of the 7th ACM conference on Recommender Systems 165\u2013172.","DOI":"10.1145\/2507157.2507163"},{"key":"e_1_2_10_17_1","unstructured":"Mnih A. andR. R.Salakhutdinov.2007.\u201cProbabilistic Matrix Factorization.\u201dAdvances in Neural Information Processing systems 20."},{"key":"e_1_2_10_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cosrev.2026.100934"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2025.129446"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.3390\/electronics13224429"},{"key":"e_1_2_10_21_1","doi-asserted-by":"crossref","unstructured":"Ren X. W.Wei L.Xia et\u00a0al.2024.\u201cRepresentation Learning With Large Language Models for Recommendation.\u201dInProceedings of the ACM Web Conference 2024 3464\u20133475.","DOI":"10.1145\/3589334.3645458"},{"key":"e_1_2_10_22_1","doi-asserted-by":"crossref","unstructured":"Sun Z. Z.Si X.Zang et\u00a0al.2024.\u201cLarge Language Models Enhanced Collaborative Filtering.\u201dInProceedings of the 33rd ACM International Conference on Information and Knowledge Management 2178\u20132188.","DOI":"10.1145\/3627673.3679558"},{"key":"e_1_2_10_23_1","doi-asserted-by":"crossref","unstructured":"Tsai A. Y. A.Kraft L.Jin et\u00a0al.2024.\u201cLeveraging Llm Reasoning Enhances Personalized Recommender Systems.\u201darXiv Preprint arXiv:2408.00802.","DOI":"10.18653\/v1\/2024.findings-acl.780"},{"issue":"8","key":"e_1_2_10_24_1","first-page":"30","article-title":"Matrix Factorization Techniques for Recommender Systems","volume":"42","author":"Volinsky C.","year":"2009","journal-title":"Computer"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i17.29887"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2022.116717"},{"key":"e_1_2_10_27_1","unstructured":"Zheng H. M.Jiang andT.Xu.2024.\u201cTowards Next\u2010Generation LLM\u2010Based Recommender Systems: A Survey and Beyond.\u201darXiv Preprint arXiv:2410.19744."},{"key":"e_1_2_10_28_1","doi-asserted-by":"crossref","unstructured":"Zheng L. V.Noroozi andP. S.Yu.2017.\u201cJoint Deep Modeling of Users and Items Using Reviews for Recommendation.\u201dIn Proceedings of the Tenth ACM International Conference on Web Search AND Data Mining 425\u2013434.","DOI":"10.1145\/3018661.3018665"}],"container-title":["Expert Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70297","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/exsy.70297","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70297","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T09:17:12Z","timestamp":1781687832000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/exsy.70297"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,15]]},"references-count":27,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1111\/exsy.70297"],"URL":"https:\/\/doi.org\/10.1111\/exsy.70297","archive":["Portico"],"relation":{},"ISSN":["0266-4720","1468-0394"],"issn-type":[{"value":"0266-4720","type":"print"},{"value":"1468-0394","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,15]]},"assertion":[{"value":"2025-11-02","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-10","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-15","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70297"}}