{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,20]],"date-time":"2026-07-20T14:01:27Z","timestamp":1784556087844,"version":"3.55.0"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"7","license":[{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,5,13]],"date-time":"2025-05-13T00:00:00Z","timestamp":1747094400000},"content-version":"vor","delay-in-days":42,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Consiglio Nazionale Delle Ricerche"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Soft Comput"],"published-print":{"date-parts":[[2025,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>In modern business environments, identifying anomalous or deviant instances in business process executions is a critical concern for enterprises and organizations. Recent advancements show that deep deviance detection models (DDMs), trained on process traces using (semi-)supervised learning techniques, outperform traditional machine learning methods. However, the effectiveness of these deep learning models often depends on large training datasets, which are not always available in practice, particularly in Green AI contexts, where data and computational resources are limited. To address these challenges, this paper presents a novel methodology for discovering deep DDMs that mitigates the impact of limited training data. Our approach incorporates an auxiliary self-supervised learning task that complements the primary deviance classification objective. In addition, we enhance the model with an autoencoder, using its reconstruction error as an additional self-supervisory signal. To promote interpretability, the model adopts a pattern-based encoding mechanism, on top of which two parallel feature-representation layers are efficiently and robustly learned through residual-like skip connections. Our method demonstrates its ability to handle the dual challenges of data efficiency and model explainability, as shown in a case study involving the execution traces of a real-world business process. The results highlight the potential of deep DDMs to achieve high performance in deviance detection, even when faced with limited data availability. Notably, our approach achieves an average performance gain (across all performance metrics) of over 15% while using only 5% of the labelled data, compared to a fully supervised baseline model, when evaluated on two publicly available logs from the current literature.<\/jats:p>","DOI":"10.1007\/s00500-025-10646-4","type":"journal-article","created":{"date-parts":[[2025,5,13]],"date-time":"2025-05-13T08:24:14Z","timestamp":1747124654000},"page":"3675-3690","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["The force of few: boosting deviance detection in data scarcity scenarios through self-supervised learning and pattern-based encoding"],"prefix":"10.1007","volume":"29","author":[{"given":"Francesco","family":"Folino","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8139-3445","authenticated-orcid":false,"given":"Gianluigi","family":"Folino","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Massimo","family":"Guarascio","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luigi","family":"Pontieri","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,5,13]]},"reference":[{"key":"10646_CR1","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1186\/s40537-021-00419-9","volume":"8","author":"A Adadi","year":"2021","unstructured":"Adadi A (2021) A survey on data-efficient algorithms in big data era. J Big Data 8:24","journal-title":"J Big Data"},{"issue":"1","key":"10646_CR2","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1002\/widm.1144","volume":"5","author":"M Atzmueller","year":"2015","unstructured":"Atzmueller M (2015) Subgroup discovery: advanced review. Wiley Int Rev Data Min Knowl Disc 5(1):35\u201349","journal-title":"Wiley Int Rev Data Min Knowl Disc"},{"key":"10646_CR3","doi-asserted-by":"crossref","unstructured":"Bose RPJC, van\u00a0der Aalst WMP (2013) Discovering signature patterns from event logs. In: IEEE symposium on computation intelligence and data mining (CIDM), Singapore. pp 111\u2013118","DOI":"10.1109\/CIDM.2013.6597225"},{"key":"10646_CR4","doi-asserted-by":"crossref","unstructured":"Cuzzocrea A, Folino F, Guarascio M, et\u00a0al (2015) A multi-view learning approach to the discovery of deviant process instances. In: Proceedings of international conference on cooperative information systems (CoopIS), Rhodes, Greece. pp 146\u2013165","DOI":"10.1007\/978-3-319-26148-5_9"},{"key":"10646_CR5","doi-asserted-by":"crossref","unstructured":"Cuzzocrea A, Folino F, Guarascio M, et\u00a0al (2016a) A multi-view multi-dimensional ensemble learning approach to mining business process deviances. In: 2016 International joint conference on neural networks (IJCNN), Vancouver, Canada. pp 3809\u20133816","DOI":"10.1109\/IJCNN.2016.7727691"},{"issue":"04","key":"10646_CR6","doi-asserted-by":"publisher","first-page":"1740003","DOI":"10.1142\/S0218843017400032","volume":"25","author":"A Cuzzocrea","year":"2016","unstructured":"Cuzzocrea A, Folino F, Guarascio M et al (2016b) A robust and versatile multi-view learning framework for the detection of deviant business process instances. Int J Coop Inf Syst 25(04):1740003","journal-title":"Int J Coop Inf Syst"},{"issue":"4","key":"10646_CR7","doi-asserted-by":"publisher","first-page":"485","DOI":"10.1109\/JPROC.2020.2976475","volume":"108","author":"L Deng","year":"2020","unstructured":"Deng L et al (2020) Model compression and hardware acceleration for neural networks: a comprehensive survey. Proc IEEE 108(4):485\u2013532","journal-title":"Proc IEEE"},{"key":"10646_CR8","doi-asserted-by":"crossref","unstructured":"Fani\u00a0Sani M, van\u00a0der Aalst W, Bolt A, et\u00a0al (2017) Subgroup discovery in process mining. In: Proceedings of 20th international conference on business information systems (BIS). Springer, pp 237","DOI":"10.1007\/978-3-319-59336-4_17"},{"key":"10646_CR9","doi-asserted-by":"crossref","unstructured":"Folino F, Pontieri L (2019) Business process deviance mining. In: Encyclopedia of big data technologies. Springer","DOI":"10.1007\/978-3-319-77525-8_100"},{"key":"10646_CR10","doi-asserted-by":"crossref","unstructured":"Folino F, Guarascio M, Pontieri L (2017) A descriptive clustering approach to the analysis of quantitative business-process deviances. In: Proceedings of 32nd ACM SIGAPP symposium on applied computing (SAC), Marrakech, Morocco, pp 765\u2013770","DOI":"10.1145\/3019612.3019660"},{"key":"10646_CR11","doi-asserted-by":"crossref","unstructured":"Folino F, Folino G, Guarascio M, et\u00a0al (2020) A multi-view ensemble of deep models for the detection of deviant process instances. In: ECML-PKDD Workshops, Ghent, Belgium, pp 249\u2013262","DOI":"10.1007\/978-3-030-65965-3_16"},{"key":"10646_CR12","doi-asserted-by":"crossref","unstructured":"Folino F, Folino G, Guarascio M, et\u00a0al (2022a) Combining active learning and fast DNN ensembles for process deviance discovery. In: 26th International symposium, ISMIS, LNCS, vol 13515. Springer, pp 346\u2013356","DOI":"10.1007\/978-3-031-16564-1_33"},{"issue":"6","key":"10646_CR13","doi-asserted-by":"publisher","first-page":"729","DOI":"10.1007\/s12599-022-00749-9","volume":"64","author":"F Folino","year":"2022","unstructured":"Folino F, Folino G, Guarascio M et al (2022b) Semi-supervised discovery of dnn-based outcome predictors from scarcely-labeled process logs. Bus Inf Syst Eng 64(6):729\u2013749","journal-title":"Bus Inf Syst Eng"},{"issue":"1","key":"10646_CR14","first-page":"10882","volume":"22","author":"T Hoefler","year":"2021","unstructured":"Hoefler T, Alistarh D, Ben-Nun T et al (2021) Sparsity in deep learning: pruning and growth for efficient inference and training in neural networks. J Mach Learn Res 22(1):10882\u201311005","journal-title":"J Mach Learn Res"},{"key":"10646_CR15","doi-asserted-by":"crossref","unstructured":"Lo D, Cheng H, Han J, et\u00a0al (2009) Classification of software behaviors for failure detection: a discriminative pattern mining approach. In: Proceedings of 15th international conference on knowledge discovery and data mining (KDD), Paris, France, pp 557\u2013566","DOI":"10.1145\/1557019.1557083"},{"key":"10646_CR16","unstructured":"Lundberg SM, Lee SI (2017) A unified approach to interpreting model predictions. In: Advances in neural information processing systems (NIPS), Long Beach, CA, USA"},{"key":"10646_CR17","doi-asserted-by":"crossref","unstructured":"Nguyen H, Dumas M, Rosa ML, et\u00a0al (2014) Mining business process deviance: a quest for accuracy. In: In Proceedings of OTM 2014 conferences, Amantea, Italy, pp 436\u2013445","DOI":"10.1007\/978-3-662-45563-0_25"},{"key":"10646_CR18","unstructured":"Nguyen H, Dumas M, La\u00a0Rosa M, et\u00a0al (2016) Business process deviance mining: review and evaluation. arXiv preprint arXiv:1608.08252"},{"key":"10646_CR19","doi-asserted-by":"crossref","unstructured":"Ribeiro MT, Singh S, Guestrin C (2016) \u201cWhy should i trust you?\" explaining the predictions of any classifier. In: Proceedings of 22nd ACM SIGKDD international conference on knowledge discovery and data mining (KDD), San Francisco, CA, USA, pp 1135\u20131144","DOI":"10.1145\/2939672.2939778"},{"key":"10646_CR20","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. In: Medical image computing and computer-assisted intervention (MICCAI), Munich, Germany, pp 234\u2013241","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"10646_CR21","doi-asserted-by":"crossref","unstructured":"Selvaraju RR, Cogswell M, Das A, et\u00a0al (2017) Grad-cam: visual explanations from deep networks via gradient-based localization. In: Proceedings of the IEEE international conference on computer vision, Venice, Italy, pp 618\u2013626","DOI":"10.1109\/ICCV.2017.74"},{"key":"10646_CR22","unstructured":"Srivastava S (2022) Towards Green AI: cost-efficient deep learning using domain knowledge. PhD thesis, The Ohio State University"},{"key":"10646_CR23","doi-asserted-by":"crossref","unstructured":"Suriadi S, Wynn MT, Ouyang C, et\u00a0al (2013) Understanding process behaviours in a large insurance company in Australia: a case study. In: Proceedings of 25th international conference on advanced information systems engineering (CAiSE), Valencia, Spain, pp 449\u2013464","DOI":"10.1007\/978-3-642-38709-8_29"},{"key":"10646_CR24","doi-asserted-by":"crossref","unstructured":"Swinnen J, Depaire B, Jans MJ, et\u00a0al (2012) A process deviation analysis\u2014a case study. In: Proceedings of international conference on business process management, Clermont-Ferrand, France, pp 87\u201398","DOI":"10.1007\/978-3-642-28108-2_8"},{"key":"10646_CR25","doi-asserted-by":"crossref","unstructured":"Van Der\u00a0Aalst W (2011) Process mining: discovery, conformance and enhancement of business processes, vol\u00a02. Berlin, Germany","DOI":"10.1007\/978-3-642-19345-3"},{"key":"10646_CR26","doi-asserted-by":"publisher","unstructured":"van Dongen B (2011) Real-life event logs-hospital log. https:\/\/doi.org\/10.4121\/uuid:d9769f3d-0ab0-4fb8-803b-0d1120ffcf54","DOI":"10.4121\/uuid:d9769f3d-0ab0-4fb8-803b-0d1120ffcf54"},{"key":"10646_CR27","doi-asserted-by":"crossref","unstructured":"Vincent P, Larochelle H, Bengio Y, et\u00a0al (2008) Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th international conference on machine learning (ICML), Helsinki, Finland, pp 1096\u20131103","DOI":"10.1145\/1390156.1390294"},{"issue":"12","key":"10646_CR28","doi-asserted-by":"publisher","first-page":"5586","DOI":"10.1109\/TKDE.2021.3070203","volume":"34","author":"Y Zhang","year":"2021","unstructured":"Zhang Y, Yang Q (2021) A survey on multi-task learning. IEEE Trans Knowl Data Eng 34(12):5586\u20135609","journal-title":"IEEE Trans Knowl Data Eng"}],"container-title":["Soft Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00500-025-10646-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00500-025-10646-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00500-025-10646-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T07:07:39Z","timestamp":1749193659000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00500-025-10646-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4]]},"references-count":28,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2025,4]]}},"alternative-id":["10646"],"URL":"https:\/\/doi.org\/10.1007\/s00500-025-10646-4","relation":{},"ISSN":["1432-7643","1433-7479"],"issn-type":[{"value":"1432-7643","type":"print"},{"value":"1433-7479","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4]]},"assertion":[{"value":"8 January 2025","order":1,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 May 2025","order":2,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors have no conflicts of interest to declare that are relevant to the content of this article.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}}]}}