{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T20:38:43Z","timestamp":1791491923808,"version":"4.3.4"},"reference-count":47,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,6,4]],"date-time":"2025-06-04T00:00:00Z","timestamp":1748995200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,6,4]],"date-time":"2025-06-04T00:00:00Z","timestamp":1748995200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001652","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001652","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Process Sci"],"abstract":"<jats:title>Abstract<\/jats:title>\n          <jats:p>Predictive business process monitoring aims to enhance process execution by providing real-time predictions about the future evolution of a process instance. In recent years, several deep learning approaches have been established as state of the art for various predictive tasks, including those based on the transformer architecture. The transformer architecture is equipped with a powerful attention mechanism that assigns attention-based importance scores to each input element, guiding the model to focus on the most relevant parts of the sequence, regardless of their position. This capability leads to more accurate and contextually grounded predictions. However, like most deep learning models, transformers largely operate as a black box, making it challenging to trace how specific features influence the model\u2019s predictions. In this paper, we conduct a series of experiments to examine the role of attention scores in a transformer-based next activity prediction model. Specifically, we investigate whether these scores provide meaningful explanations for the model\u2019s decisions. Our findings reveal that attention scores can indeed serve as effective explanations. Building on these insights, we propose two novel, global, graph-based explanation approaches that illustrate the model\u2019s understanding of the process\u2019s control flow. Our evaluation using various metrics on both real-world and synthetic event logs demonstrates that these explainers effectively capture the model\u2019s decision-making process. By improving interpretability, these insights not only enhance process participants\u2019 confidence in predictive models but also offer a valuable foundation for refining model performance. Furthermore, our investigation into the reliability of attention scores offers valuable insights into how transformer models encapsulate temporal and sequential dependencies in prediction tasks.<\/jats:p>","DOI":"10.1007\/s44311-025-00018-4","type":"journal-article","created":{"date-parts":[[2025,6,4]],"date-time":"2025-06-04T15:35:34Z","timestamp":1749051334000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":6,"title":["Explaining transformer-based next activity prediction by using attention scores"],"prefix":"10.1007","volume":"2","author":[{"given":"Martin","family":"K\u00e4ppel","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lars","family":"Ackermann","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefan","family":"Jablonski","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Simon","family":"H\u00e4rtl","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,6,4]]},"reference":[{"key":"18_CR1","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1016\/j.tjem.2018.08.001","volume":"18","author":"H Ako\u011flu","year":"2018","unstructured":"Ako\u011flu H (2018) User\u2019s guide to correlation coefficients. Turk J Emerg Med 18:91\u201393","journal-title":"Turk J Emerg Med"},{"key":"18_CR2","doi-asserted-by":"publisher","unstructured":"Bukhsh ZA, Saeed A, Dijkman RM (2021) Processtransformer: Predictive business process monitoring with transformer network.\u00a0arXiv. https:\/\/doi.org\/10.48550\/arXiv.2104.00721","DOI":"10.48550\/arXiv.2104.00721"},{"key":"18_CR3","unstructured":"Burattin A, Re B, Rossi L et\u00a0al (2022) Purple: a purpose-guided log generator. In: Proceedings of the ICPM Doctoral Consortium and Demo Track 2022. CEUR-WS, CEUR Workshop Proceedings. iCPM 2022 Demo Track; Conference date: 23-10-2022 Through 28-10-2022.\u00a0https:\/\/icpmconference.org\/2022\/.\u00a0Access Dates: 17.05.2025"},{"key":"18_CR4","doi-asserted-by":"crossref","unstructured":"Camargo M, Dumas M, Gonz\u00e1lez-Rojas O (2019) Learning accurate lstm models of business processes. In: Proc. of BPM Conf. Springer, Cham, pp 286\u2013302","DOI":"10.1007\/978-3-030-26619-6_19"},{"issue":"8","key":"18_CR5","doi-asserted-by":"publisher","first-page":"832","DOI":"10.3390\/electronics8080832","volume":"8","author":"DV Carvalho","year":"2019","unstructured":"Carvalho DV, Pereira EM, Cardoso JS (2019) Machine Learning Interpretability: A Survey on Methods and Metrics. Electronics 8(8):832. https:\/\/doi.org\/10.3390\/electronics8080832","journal-title":"Electronics"},{"key":"18_CR6","doi-asserted-by":"publisher","unstructured":"Devlin J, Chang MW, Lee K et\u00a0al (2019) BERT: Pre-training of deep bidirectional transformers for language understanding. In: Burstein J, Doran C, Solorio T (eds) Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). Association for Computational Linguistics, Minneapolis, pp 4171\u20134186. https:\/\/doi.org\/10.18653\/v1\/N19-1423. https:\/\/aclanthology.org\/N19-1423\/","DOI":"10.18653\/v1\/N19-1423"},{"key":"18_CR7","first-page":"320","volume-title":"Predictive Process Monitoring","author":"C Di Francescomarino","year":"2022","unstructured":"Di Francescomarino C, Ghidini C (2022) Predictive Process Monitoring. Springer, Cham, pp 320\u2013346"},{"key":"18_CR8","doi-asserted-by":"crossref","unstructured":"Dumas M, Rosa ML, Mendling J et al (2018) Fundamentals of Business Process Management, 2nd edn. Springer Publishing Company, Incorporated,\u00a0Berlin.","DOI":"10.1007\/978-3-662-56509-4"},{"key":"18_CR9","doi-asserted-by":"publisher","first-page":"8192","DOI":"10.3390\/app12168192","volume":"12","author":"G Elkhawaga","year":"2022","unstructured":"Elkhawaga G, Abu-Elkheir M, Reichert M (2022) Explainability of Predictive Process Monitoring Results: Can You See My Data Issues? Appl Sci 12:8192. https:\/\/doi.org\/10.3390\/app12168192","journal-title":"Appl Sci"},{"key":"18_CR10","doi-asserted-by":"crossref","unstructured":"Elkhawaga G, Abu-Elkheir M, Reichert M (2022) Explainability of predictive process monitoring results: Can you see my data issues? Appl Sci 12(16)","DOI":"10.3390\/app12168192"},{"key":"18_CR11","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1016\/j.dss.2017.04.003","volume":"100","author":"J Evermann","year":"2017","unstructured":"Evermann J, Rehse JR, Fettke P (2017) Predicting process behaviour using deep learning. Decis Support Syst 100:129\u2013140","journal-title":"Decis Support Syst"},{"issue":"3","key":"18_CR12","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1016\/j.compind.2003.10.007","volume":"53","author":"D Grigori","year":"2004","unstructured":"Grigori D, Casati F, Castellanos M et al (2004) Business process intelligence. Comput Ind 53(3):321\u2013343","journal-title":"Comput Ind"},{"key":"18_CR13","doi-asserted-by":"publisher","unstructured":"Guidotti, R., Monreale, A., Ruggieri, S., Pedreschi, D., Turini, F. and Giannotti, F. (2018) Local Rule-Based Explanations of Black Box Decision Systems. arXiv Preprint arXiv: 1805.10820. https:\/\/doi.org\/10.48550\/arXiv.1805.10820","DOI":"10.48550\/arXiv.1805.10820"},{"key":"18_CR14","doi-asserted-by":"publisher","first-page":"172923","DOI":"10.1109\/ACCESS.2020.3025999","volume":"8","author":"KM Hanga","year":"2020","unstructured":"Hanga KM, Kovalchuk Y, Gaber MM (2020) A graph-based approach to interpreting recurrent neural networks in process mining. IEEE Access 8:172923\u2013172938","journal-title":"IEEE Access"},{"key":"18_CR15","doi-asserted-by":"crossref","unstructured":"Huang TH, Metzger A, Pohl K (2022) Counterfactual explanations for predictive business process monitoring. In: Information Systems. Springer, Cham","DOI":"10.1007\/978-3-030-95947-0_28"},{"key":"18_CR16","unstructured":"Jain S, Wallace BC (2019) Attention is not Explanation. In: Proc. of 2019 Conf. of the North American Chapter of the ACL). ACL, Minneapolis"},{"key":"18_CR17","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1007\/978-3-031-70396-6_12","volume-title":"Business Process Management","author":"M K\u00e4ppel","year":"2024","unstructured":"K\u00e4ppel M, Ackermann L, Jablonski S et al (2024) Attention please: What transformer models really learn for process prediction. In: Marrella A, Resinas M, Jans M et al (eds) Business Process Management. Springer Nature Switzerland, Cham, pp 203\u2013220"},{"key":"18_CR18","doi-asserted-by":"publisher","unstructured":"K\u00e4ppel, M., Jablonski, S., Sch\u00f6nig, S. (2021). Evaluating Predictive Business Process Monitoring Approaches on Small Event Logs. In: Paiva, A.C.R., Cavalli, A.R., Ventura Martins, P., P\u00e9rez-Castillo, R. (eds) Quality of Information and Communications Technology. QUATIC 2021. Communications in Computer and Information Science, Springer, Cham, vol 1439:167\u2013182. https:\/\/doi.org\/10.1007\/978-3-030-85347-1_13","DOI":"10.1007\/978-3-030-85347-1_13"},{"issue":"1\/2","key":"18_CR19","doi-asserted-by":"publisher","first-page":"81","DOI":"10.2307\/2332226","volume":"30","author":"MG Kendall","year":"1938","unstructured":"Kendall MG (1938) A new measure of rank correlation. Biometrika 30(1\/2):81\u201393","journal-title":"Biometrika"},{"key":"18_CR20","doi-asserted-by":"publisher","unstructured":"Klinkm\u00fcller C, van Beest NRTP, Weber I (2018) Towards reliable predictive process monitoring. In: Information Systems in the Big Data Era. Springer,\u00a0Cham 317:163\u2013181. https:\/\/doi.org\/10.1007\/978-3-319-92901-9_15","DOI":"10.1007\/978-3-319-92901-9_15"},{"issue":"3","key":"18_CR21","first-page":"261","volume":"63","author":"W Kratsch","year":"2021","unstructured":"Kratsch W, Manderscheid J, R\u00f6glinger M et al (2021) Machine learning in business process monitoring: A comparison of deep learning and classical approaches used for outcome prediction. BISE 63(3):261\u2013276","journal-title":"BISE"},{"issue":"1","key":"18_CR22","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1214\/aoms\/1177729487","volume":"23","author":"S Kullback","year":"1952","unstructured":"Kullback S (1952) An application of information theory to multivariate analysis. Ann Math Stat 23(1):88\u2013102","journal-title":"Ann Math Stat"},{"key":"18_CR23","doi-asserted-by":"publisher","unstructured":"Maggi FM, Di\u00a0Francescomarino C, Dumas M et\u00a0al (2014) Predictive monitoring of business processes. In: Proceedings of the 26th International Conference on Advanced Information Systems Engineering. Springer,\u00a0Cham, vol\u00a08484:pp 457\u2013472. https:\/\/doi.org\/10.1007\/978-3-319-07881-6_31","DOI":"10.1007\/978-3-319-07881-6_31"},{"key":"18_CR24","doi-asserted-by":"crossref","unstructured":"Manning C, Raghavan P, Sch\u00fctze H (2008) Introduction to Information Retrieval. Cambridge University Press,\u00a0Cambridge.","DOI":"10.1017\/CBO9780511809071"},{"issue":"6","key":"18_CR25","doi-asserted-by":"publisher","first-page":"962","DOI":"10.1109\/TSC.2017.2772256","volume":"11","author":"AE M\u00e1rquez-Chamorro","year":"2017","unstructured":"M\u00e1rquez-Chamorro AE, Resinas M, Ruiz-Cort\u00e9s A (2017) Predictive monitoring of business processes: A survey. IEEE Trans Serv Comput 11(6):962\u2013977","journal-title":"IEEE Trans Serv Comput"},{"key":"18_CR26","doi-asserted-by":"publisher","first-page":"143","DOI":"10.1007\/s12599-018-0551-3","volume":"62","author":"N Mehdiyev","year":"2020","unstructured":"Mehdiyev N, Evermann J, Fettke P (2020) A novel business process prediction model using a deep learning method. Bus Inf Syst Eng 62:143\u2013157","journal-title":"Bus Inf Syst Eng"},{"key":"18_CR27","doi-asserted-by":"publisher","unstructured":"Mehdiyev, N., Fettke, P. (2021). Explainable Artificial Intelligence for Process Mining: A General Overview and Application of a Novel Local Explanation Approach for Predictive Process Monitoring. In: Pedrycz, W., Chen, SM. (eds) Interpretable Artificial Intelligence: A Perspective of Granular Computing. Studies in Computational Intelligence, Springer, Cham, vol 937:1\u201328.\u00a0https:\/\/doi.org\/10.1007\/978-3-030-64949-4_1","DOI":"10.1007\/978-3-030-64949-4_1"},{"key":"18_CR28","doi-asserted-by":"publisher","unstructured":"Nagahisarchoghaei M, Nur N, Cummins L et\u00a0al (2023) An empirical survey on explainable ai technologies: Recent trends, use-cases, and categories from technical and application perspectives. Electronics 12(5). https:\/\/doi.org\/10.3390\/electronics12051092","DOI":"10.3390\/electronics12051092"},{"key":"18_CR29","doi-asserted-by":"crossref","unstructured":"Nauta M, Trienes J, Pathak S et\u00a0al (2023) From anecdotal evidence to quantitative evaluation methods: A systematic review on evaluating explainable ai. ACM 55(13s)","DOI":"10.1145\/3583558"},{"key":"18_CR30","unstructured":"OpenAI, Achiam J, Adler S et\u00a0al (2024) Gpt-4 technical report. https:\/\/arxiv.org\/abs\/2303.08774"},{"key":"18_CR31","doi-asserted-by":"publisher","unstructured":"Pasquadibisceglie V, Appice A, Castellano G et\u00a0al (2019) Using convolutional neural networks for predictive process analytics. In:\u00a02019 International Conference on Process Mining (ICPM), IEEE,\u00a0 129\u2013136.\u00a0https:\/\/doi.org\/10.1109\/ICPM.2019.00028","DOI":"10.1109\/ICPM.2019.00028"},{"issue":"1","key":"18_CR32","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1007\/s10844-022-00765-x","volume":"61","author":"J Peeperkorn","year":"2023","unstructured":"Peeperkorn J, Sv Broucke, De Weerdt J (2023) Can recurrent neural networks learn process model structure? J Intell Inf Syst 61(1):27\u201351","journal-title":"J Intell Inf Syst"},{"issue":"1","key":"18_CR33","first-page":"739","volume":"16","author":"E Rama-Maneiro","year":"2021","unstructured":"Rama-Maneiro E, Vidal JC, Lama M (2021) Deep learning for predictive business process monitoring: Review and benchmark. IEEE Trans Serv Comput 16(1):739\u2013756","journal-title":"IEEE Trans Serv Comput"},{"key":"18_CR34","first-page":"141","volume-title":"BPM Forum","author":"W Rizzi","year":"2020","unstructured":"Rizzi W, Di Francescomarino C, Maggi FM (2020) Explainability in predictive process monitoring: When understanding helps improving. BPM Forum. Springer, Cham, pp 141\u2013158"},{"key":"18_CR35","unstructured":"Rojat T, Puget R, Filliat D et\u00a0al (2021) Explainable artificial intelligence (xai) on timeseries data: A survey.\u00a0https:\/\/arxiv.org\/abs\/2104.00950"},{"key":"18_CR36","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1016\/j.is.2019.01.007","volume":"84","author":"A Senderovich","year":"2019","unstructured":"Senderovich A, Di Francescomarino C, Maggi FM (2019) From knowledge-driven to data-driven inter-case feature encoding in predictive process monitoring. Inf Syst 84:255\u2013264","journal-title":"Inf Syst"},{"key":"18_CR37","doi-asserted-by":"crossref","unstructured":"Serrano S, Smith NA (2019) Is attention interpretable? In: Proc. of the 57th Annual Meeting of the ACL. ACL, Florence, pp 2931\u20132951","DOI":"10.18653\/v1\/P19-1282"},{"key":"18_CR38","doi-asserted-by":"crossref","unstructured":"Sindhgatta R, Moreira C, Ouyang C et al (2020) Exploring interpretable predictive models for business processes. Springer","DOI":"10.1007\/978-3-030-58666-9_15"},{"key":"18_CR39","doi-asserted-by":"publisher","first-page":"194","DOI":"10.1007\/978-3-030-98581-3_15","volume-title":"Process Mining Workshops","author":"A Stevens","year":"2022","unstructured":"Stevens A, De Smedt J, Peeperkorn J (2022) Quantifying explainability in outcome-oriented predictive process monitoring. Process Mining Workshops. Springer, Cham, pp 194\u2013206"},{"key":"18_CR40","unstructured":"Stierle, Matthias; Brunk, Jens; Weinzierl, Sven; Zilker, Sandra; Matzner, Martin; and Becker, J\u00f6rg, Bringing Light Into the Darkness - A Systematic Literature Review on Explainable Predictive Business Process Monitoring Techniques (2021). ECIS 2021 Research-in-Progress Papers, 8. https:\/\/aisel.aisnet.org\/ecis2021_rip\/8"},{"key":"18_CR41","doi-asserted-by":"publisher","unstructured":"van der Aalst WMP (2016) Process Mining. Springer, Berlin.\u00a0https:\/\/doi.org\/10.1007\/978-3-662-49851-4","DOI":"10.1007\/978-3-662-49851-4"},{"key":"18_CR42","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, u. & Polosukhin, I. (2017). Attention is All you Need. In I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan & R. Garnett (eds.), Advances in Neural Information Processing Systems, : Curran Associates, Inc 5998\u20136008.\u00a0http:\/\/arxiv.org\/abs\/1706.03762"},{"issue":"4","key":"18_CR43","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3331449","volume":"10","author":"I Verenich","year":"2019","unstructured":"Verenich I, Dumas M, Rosa ML et al (2019) Survey and cross-benchmark comparison of remaining time prediction methods in business process monitoring. ACM Trans Intell Syst Technol 10(4):1\u201334. https:\/\/doi.org\/10.1145\/3331449","journal-title":"ACM Trans Intell Syst Technol"},{"key":"18_CR44","doi-asserted-by":"publisher","first-page":"124181","DOI":"10.1016\/j.eswa.2024.124181","volume":"253","author":"S Weinzierl","year":"2024","unstructured":"Weinzierl S, Zilker S, Dunzer S et al (2024) Machine learning in business process management: A systematic literature review. Exp Syst Appl 253:124181. https:\/\/doi.org\/10.1016\/j.eswa.2024.124181","journal-title":"Exp Syst Appl"},{"key":"18_CR45","doi-asserted-by":"publisher","unstructured":"Weinzierl, S., Zilker, S., Brunk, J., Revoredo, K., Matzner, M., Becker, J. (2020). XNAP: Making LSTM-Based Next Activity Predictions Explainable by Using LRP. In: Del R\u00edo Ortega, A., Leopold, H., Santoro, F.M. (eds) Business Process Management Workshops. BPM 2020. Lecture Notes in Business Information Processing, Springer, Cham.\u00a0vol 397:\u00a0129\u2013141.\u00a0https:\/\/doi.org\/10.1007\/978-3-030-66498-5_10","DOI":"10.1007\/978-3-030-66498-5_10"},{"key":"18_CR46","doi-asserted-by":"publisher","unstructured":"Wiegreffe S, Pinter Y (2019) Attention is not not explanation. Inui K, Jiang J, Ng V and Wan X (eds),In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP),\u00a0Association for Computational Linguistics, Hong Kong,\u00a011\u201320,\u00a0https:\/\/doi.org\/10.18653\/v1\/D19-1002","DOI":"10.18653\/v1\/D19-1002"},{"key":"18_CR47","unstructured":"Zhao WX, Zhou K, Li J et\u00a0al (2025) A survey of large language models.\u00a0https:\/\/arxiv.org\/abs\/2303.18223"}],"container-title":["Process Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s44311-025-00018-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s44311-025-00018-4\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s44311-025-00018-4.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,10]],"date-time":"2025-07-10T09:09:50Z","timestamp":1752138590000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s44311-025-00018-4"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,4]]},"references-count":47,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["18"],"URL":"https:\/\/doi.org\/10.1007\/s44311-025-00018-4","relation":{},"ISSN":["2948-2178"],"issn-type":[{"value":"2948-2178","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,4]]},"assertion":[{"value":"9 November 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 May 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 June 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to paticipate"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"11"}}