{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,12]],"date-time":"2025-12-12T08:45:55Z","timestamp":1765529155174,"version":"3.48.0"},"reference-count":40,"publisher":"Springer Science and Business Media LLC","issue":"4","license":[{"start":{"date-parts":[[2025,7,14]],"date-time":"2025-07-14T00:00:00Z","timestamp":1752451200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,7,14]],"date-time":"2025-07-14T00:00:00Z","timestamp":1752451200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["No.62406057"],"award-info":[{"award-number":["No.62406057"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100018537","name":"National Science and Technology Major Project","doi-asserted-by":"publisher","award":["2023ZD0501806"],"award-info":[{"award-number":["2023ZD0501806"]}],"id":[{"id":"10.13039\/501100018537","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Data Sci. Eng."],"published-print":{"date-parts":[[2025,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Session-based recommendation (SBR) focuses on predicting the next potential item for anonymous users based on short-click sessions. However, these interaction sessions often contain noise items, which arise from misclicks or shifts in user interests. Existing denoising methods typically presume a strong exclusionary relationship between noise items and the recommendation target, assuming that reducing noise can enhance recommendation accuracy. In contrast, our observations reveal a nuanced phenomenon: as the length of the interaction session shortens, the effect of noise removal on recommendation performance gradually transitions from positive to negative. This finding suggests that in short sessions with insufficient contextual information, relying solely on the exclusion of noise items within the session may fail to improve\u00a0and could even hinder-the recommendation performance. Such complexities have been largely overlooked in prior research. To bridge this gap, we propose two solutions: (i) expanding the view of denoising from a single session to multiple sessions (i.e., from local to global), and (ii) introducing relevant contextual information into each session by employing enhancement strategies. Therefore, we design the Hybrid Prototype-based In-and-Out Flow Network (HyPro), which employs both denoising and enhancing processes for each session based on our proposed hybrid prototypes. Specifically, for each item, HyPro first learns the hybrid prototype by aggregating information from the item\u2019s semantic and topological neighbors across all sessions. Then, based on the hybrid prototypes, HyPro employs an in-and-out flow network comprising two components: (i) the out-flow channel, which targets the removal of irrelevant information at both the data and feature levels, and (ii) the in-flow channel, which integrates global information for each session at the item and session levels. Extensive experiments conducted on three real-world datasets demonstrate that HyPro outperforms the state-of-the-art baselines. The implementation code is available at:\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/jarviswww\/Code4HyPro\" ext-link-type=\"uri\">https:\/\/github.com\/jarviswww\/Code4HyPro<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1007\/s41019-025-00301-1","type":"journal-article","created":{"date-parts":[[2025,7,14]],"date-time":"2025-07-14T04:42:53Z","timestamp":1752468173000},"page":"711-728","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Rethinking the Denoising Strategy in Session-Based Recommendation via Bidirectional Information Flow"],"prefix":"10.1007","volume":"10","author":[{"given":"Xiao","family":"Wang","sequence":"first","affiliation":[]},{"given":"Tingting","family":"Dai","sequence":"additional","affiliation":[]},{"given":"Wudong","family":"Cai","sequence":"additional","affiliation":[]},{"given":"Ke","family":"Qin","sequence":"additional","affiliation":[]},{"given":"Jie","family":"Shao","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7387-2801","authenticated-orcid":false,"given":"Shuang","family":"Liang","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2025,7,14]]},"reference":[{"key":"301_CR1","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1016\/j.knosys.2013.03.012","volume":"46","author":"J Bobadilla","year":"2013","unstructured":"Bobadilla J, Ortega F, Hernando A, Guti\u00e9rrez A (2013) Recommender systems survey. Knowl Based Syst 46:109\u2013132","journal-title":"Knowl Based Syst"},{"issue":"7","key":"301_CR2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3465401","volume":"54","author":"S Wang","year":"2021","unstructured":"Wang S, Cao L, Wang Y, Sheng QZ, Orgun MA, Lian D (2021) A survey on session-based recommender systems. ACM Comput Surv 54(7):1\u201338","journal-title":"ACM Comput Surv"},{"issue":"1","key":"301_CR3","doi-asserted-by":"publisher","first-page":"212","DOI":"10.1109\/TBDATA.2022.3154778","volume":"9","author":"Z Wang","year":"2022","unstructured":"Wang Z, Wang Z, Li X, Yu Z, Guo B, Chen L, Zhou X (2022) Exploring multi-dimension user-item interactions with attentional knowledge graph neural networks for recommendation. IEEE Trans Big Data 9(1):212\u2013226","journal-title":"IEEE Trans Big Data"},{"issue":"1","key":"301_CR4","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1007\/s11257-020-09277-1","volume":"31","author":"M Ludewig","year":"2021","unstructured":"Ludewig M, Mauro N, Latifi S, Jannach D (2021) Empirical analysis of session-based recommendation algorithms: a comparison of neural and non-neural approaches. User Model User Adapt Interact 31(1):149\u2013181","journal-title":"User Model User Adapt Interact"},{"key":"301_CR5","doi-asserted-by":"crossref","unstructured":"Wang S, Zhang Q, Hu L, Zhang X, Wang Y, Aggarwal C (2022) Sequential\/session-based recommendations: challenges, approaches, applications and opportunities. In: SIGIR, ACM, Madrid, Spain, pp 3425\u20133428","DOI":"10.1145\/3477495.3532685"},{"key":"301_CR6","first-page":"453","volume":"6","author":"G Shani","year":"2005","unstructured":"Shani G, Heckerman D, Brafman RI, Boutilier C (2005) An MDP-based recommender system. J Mach Learn Res 6:453\u2013460","journal-title":"J Mach Learn Res"},{"key":"301_CR7","doi-asserted-by":"crossref","unstructured":"Rendle S, Freudenthaler C, Schmidt-Thieme L (2010) Factorizing personalized Markov chains for next-basket recommendation. In: WWW, ACM, New York, NY, USA, pp 811\u2013820","DOI":"10.1145\/1772690.1772773"},{"key":"301_CR8","doi-asserted-by":"crossref","unstructured":"Hidasi B, Karatzoglou A (2018) Recurrent neural networks with top-k gains for session-based recommendations. In: CIKM, Association for Computing Machinery, New York, NY, USA, pp 843\u2013852","DOI":"10.1145\/3269206.3271761"},{"key":"301_CR9","doi-asserted-by":"crossref","unstructured":"Wang M, Ren P, Mei L, Chen Z, Ma J, Rijke M (2019) A collaborative session-based recommendation approach with parallel memory modules. In: SIGIR, ACM, Paris,France, pp 345\u201335","DOI":"10.1145\/3331184.3331210"},{"key":"301_CR10","doi-asserted-by":"crossref","unstructured":"Li J, Ren P, Chen Z, Ren Z, Lian T, Ma J (2017) Neural attentive session-based recommendation. In: CIKM, ACM, New York, NY, United States, pp 1419\u20131428","DOI":"10.1145\/3132847.3132926"},{"key":"301_CR11","doi-asserted-by":"crossref","unstructured":"Kang W-C, McAuley J (2018) Self-attentive sequential recommendation. In: ICDM, IEEE, Singapore, Singapore, pp 197\u2013206","DOI":"10.1109\/ICDM.2018.00035"},{"key":"301_CR12","doi-asserted-by":"crossref","unstructured":"Liu Q, Zeng Y, Mokhosi R, Zhang H (2018) Stamp: short-term attention\/memory priority model for session-based recommendation. In: SIGKDD, ACM, New York, NY, USA, pp 1831\u20131839","DOI":"10.1145\/3219819.3219950"},{"key":"301_CR13","doi-asserted-by":"crossref","unstructured":"Ren P, Chen Z, Li J, Ren Z, Ma J, Rijke M (2019) Repeatnet: a repeat aware neural recommendation machine for session-based recommendation. In: AAAI, AAAI Press, Honolulu, Hawaii, USA, pp 4806\u20134813","DOI":"10.1609\/aaai.v33i01.33014806"},{"key":"301_CR14","doi-asserted-by":"crossref","unstructured":"Wu S, Tang Y, Zhu Y, Wang L, Xie X, Tan T (2019) Session-based recommendation with graph neural networks. In: AAAI, Honolulu, Hawaii, USA, vol 33, pp 346\u2013353","DOI":"10.1609\/aaai.v33i01.3301346"},{"key":"301_CR15","doi-asserted-by":"crossref","unstructured":"Wang Z, Wei W, Cong G, Li X-L, Mao X-L, Qiu M (2020) Global context enhanced graph neural networks for session-based recommendation. In: SIGIR, ACM, New York, NY, USA, pp 169\u2013178","DOI":"10.1145\/3397271.3401142"},{"key":"301_CR16","doi-asserted-by":"crossref","unstructured":"Pan Z, Cai F, Chen W, Chen H, De\u00a0Rijke M (2020) Star graph neural networks for session-based recommendation. In: CIKM, ACM, Virtual Event Ireland, pp 1195\u20131204","DOI":"10.1145\/3340531.3412014"},{"key":"301_CR17","doi-asserted-by":"publisher","first-page":"4503","DOI":"10.1609\/aaai.v35i5.16578","volume":"35","author":"X Xia","year":"2021","unstructured":"Xia X, Yin H, Yu J, Wang Q, Cui L, Zhang X (2021) Self-supervised hypergraph convolutional networks for session-based recommendation. AAAI 35:4503\u20134511","journal-title":"AAAI"},{"key":"301_CR18","doi-asserted-by":"crossref","unstructured":"Xia X, Yin H, Yu J, Shao Y, Cui L (2021) Self-supervised graph co-training for session-based recommendation. In: CIKM, ACM, New York, NY, USA, pp 2180\u20132190","DOI":"10.1145\/3459637.3482388"},{"key":"301_CR19","doi-asserted-by":"crossref","unstructured":"Zhang P, Guo J, Li C, Xie Y, Kim J, Zhang Y, Xie X, Wang H, Kim S (2023) Efficiently leveraging multi-level user intent for session-based recommendation via atten-mixer network. In: WSDM, ACM, Singapore, pp 168\u2013176","DOI":"10.1145\/3539597.3570445"},{"key":"301_CR20","doi-asserted-by":"crossref","unstructured":"Chen Q, Guo Z, Li J, Li G (2023) Knowledge-enhanced multi-view graph neural networks for session-based recommendation. In: SIGIR, ACM, Taipei,Taiwan, pp 352\u2013361","DOI":"10.1145\/3539618.3591706"},{"key":"301_CR21","doi-asserted-by":"crossref","unstructured":"Wu H, Fang H, Sun Z, Geng C, Kong X, Ong Y (2023) A generic reinforced explainable framework with knowledge graph for session-based recommendation. In: ICDE, IEEE, Anaheim, CA, USA, pp 1260\u20131272","DOI":"10.1109\/ICDE55515.2023.00101"},{"key":"301_CR22","unstructured":"Wang X, Dai T, Liu Q, Liang S (2024) Spatial-temporal perceiving: deciphering user hierarchical intent in session-based recommendation. In: IJCAI, ijcai.org, Jeju, South Korea, pp 2415\u20132423"},{"key":"301_CR23","doi-asserted-by":"crossref","unstructured":"Zhang X, Li B, Jin B (2024) Denoising long- and short-term interests for sequential recommendation. In: SDM, SIAM, Houston, TX, USA, pp 544\u2013552","DOI":"10.1137\/1.9781611978032.63"},{"key":"301_CR24","doi-asserted-by":"crossref","unstructured":"Jeong J, Choi J, Cho H, Chung S (2022) Fpadametric: false-positive-aware adaptive metric learning for session-based recommendation. In: AAAI, AAAI Press, Virtual Event, pp 4039\u20134047","DOI":"10.1609\/aaai.v36i4.20321"},{"key":"301_CR25","doi-asserted-by":"crossref","unstructured":"Lin Y, Wang C, Chen Z, Ren Z, Xin X, Yan Q, Rijke M, Cheng X, Ren P (2023) A self-correcting sequential recommender. In: WWW, ACM, Austin, TX, USA, pp 1283\u20131293","DOI":"10.1145\/3543507.3583479"},{"key":"301_CR26","doi-asserted-by":"crossref","unstructured":"Ren X, Xia L, Zhao J, Yin D, Huang C (2023) Disentangled contrastive collaborative filtering. In: SIGIR, ACM, Taipei,Taiwan, pp 1137\u20131146","DOI":"10.1145\/3539618.3591665"},{"key":"301_CR27","doi-asserted-by":"publisher","first-page":"117391","DOI":"10.1016\/j.eswa.2022.117391","volume":"203","author":"C Zhang","year":"2022","unstructured":"Zhang C, Zheng W, Liu Q, Nie J, Zhang H (2022) SEDGN: sequence enhanced denoising graph neural network for session-based recommendation. Expert Syst Appl 203:117391","journal-title":"Expert Syst Appl"},{"key":"301_CR28","doi-asserted-by":"publisher","first-page":"4635","DOI":"10.1609\/aaai.v35i5.16593","volume":"35","author":"J Yuan","year":"2021","unstructured":"Yuan J, Song Z, Sun M, Wang X, Zhao WX (2021) Dual sparse attention network for session-based recommendation. AAAI 35:4635\u20134643","journal-title":"AAAI"},{"key":"301_CR29","doi-asserted-by":"crossref","unstructured":"Dai J, Yuan W, Bao C, Zhang Z (2022) DGNN: denoising graph neural network for session-based recommendation. In: DSAA, IEEE, Shenzhen, China, pp 1\u20138","DOI":"10.1109\/DSAA54385.2022.10032399"},{"key":"301_CR30","doi-asserted-by":"publisher","first-page":"123845","DOI":"10.1016\/j.eswa.2024.123845","volume":"249","author":"Z Luo","year":"2024","unstructured":"Luo Z, Sheng Z, Zhang T (2024) Dual perspective denoising model for session-based recommendation. Expert Syst Appl 249:123845","journal-title":"Expert Syst Appl"},{"issue":"4","key":"301_CR31","doi-asserted-by":"publisher","first-page":"535","DOI":"10.3390\/e24040535","volume":"24","author":"Y Yan","year":"2022","unstructured":"Yan Y, Yu G, Yan X (2022) Entropy-enhanced attention model for explanation recommendation. Entropy 24(4):535","journal-title":"Entropy"},{"issue":"2","key":"301_CR32","doi-asserted-by":"publisher","first-page":"402","DOI":"10.3390\/electronics12020402","volume":"12","author":"Y Yuan","year":"2023","unstructured":"Yuan Y, Chen L, Yang J (2023) A multidimensional model for recommendation systems based on classification and entropy. Electronics 12(2):402","journal-title":"Electronics"},{"key":"301_CR33","doi-asserted-by":"crossref","unstructured":"Tishby N, Zaslavsky N (2015) Deep learning and the information bottleneck principle. In: ITW, IEEE, Jerusalem, Israel, pp 1\u20135","DOI":"10.1109\/ITW.2015.7133169"},{"key":"301_CR34","doi-asserted-by":"crossref","unstructured":"Saxe AM, Bansal Y, Dapello J, Advani M, Kolchinsky A, Tracey BD, Cox DD (2018) On the information bottleneck theory of deep learning. In: ICLR. OpenReview.net, Vancouver, BC, Canada","DOI":"10.1088\/1742-5468\/ab3985"},{"key":"301_CR35","unstructured":"Tishby N, Pereira FC, Bialek W (2000) The information bottleneck method. arXiv preprint physics\/0004057"},{"key":"301_CR36","unstructured":"Hidasi B, Karatzoglou A, Baltrunas L, Tikk D (2016) Session-based recommendations with recurrent neural networks. In: ICLR, San Juan, Puerto Rico"},{"key":"301_CR37","doi-asserted-by":"publisher","first-page":"109911","DOI":"10.1016\/j.patcog.2023.109911","volume":"145","author":"Z Wang","year":"2024","unstructured":"Wang Z, Wei W, Zou D, Liu Y, Li X, Mao X, Qiu M (2024) Exploring global information for session-based recommendation. Pattern Recognit 145:109911","journal-title":"Pattern Recognit"},{"key":"301_CR38","doi-asserted-by":"crossref","unstructured":"Chen H, Lin Y, Pan M, Wang L, Yeh CM, Li X, Zheng Y, Wang F, Yang H (2022) Denoising self-attentive sequential recommendation. In: RecSys, ACM, Seattle, WA, USA, pp 92\u2013101","DOI":"10.1145\/3523227.3546788"},{"key":"301_CR39","doi-asserted-by":"crossref","unstructured":"Peters B, Niculae V, Martins AFT (2019) Sparse sequence-to-sequence models. In: ACL, Association for Computational Linguistics, Florence, Italy, pp 1504\u20131519","DOI":"10.18653\/v1\/P19-1146"},{"key":"301_CR40","unstructured":"Gupta P, Garg D, Malhotra P, Vig L, Shroff GM (2019) Niser: normalized item and session representations with graph neural networks. ArXiv"}],"container-title":["Data Science and Engineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s41019-025-00301-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s41019-025-00301-1\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s41019-025-00301-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,12]],"date-time":"2025-12-12T08:41:51Z","timestamp":1765528911000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s41019-025-00301-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,14]]},"references-count":40,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,12]]}},"alternative-id":["301"],"URL":"https:\/\/doi.org\/10.1007\/s41019-025-00301-1","relation":{},"ISSN":["2364-1185","2364-1541"],"issn-type":[{"type":"print","value":"2364-1185"},{"type":"electronic","value":"2364-1541"}],"subject":[],"published":{"date-parts":[[2025,7,14]]},"assertion":[{"value":"26 March 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"30 April 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 May 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 July 2025","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no Conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}