A Novel Business Process Prediction Model Using a Deep Learning Method

  • Nijat MehdiyevEmail author
  • Joerg Evermann
  • Peter Fettke
Research Paper


The ability to proactively monitor business processes is a main competitive differentiator for firms. Process execution logs generated by process aware information systems help to make process specific predictions for enabling a proactive situational awareness. The goal of the proposed approach is to predict the next process event from the completed activities of the running process instance, based on the execution log data from previously completed process instances. By predicting process events, companies can initiate timely interventions to address undesired deviations from the desired workflow. The paper proposes a multi-stage deep learning approach that formulates the next event prediction problem as a classification problem. Following a feature pre-processing stage with n-grams and feature hashing, a deep learning model consisting of an unsupervised pre-training component with stacked autoencoders and a supervised fine-tuning component is applied. Experiments on a variety of business process log datasets show that the multi-stage deep learning approach provides promising results. The study also compared the results to existing deep recurrent neural networks and conventional classification approaches. Furthermore, the paper addresses the identification of suitable hyperparameters for the proposed approach, and the handling of the imbalanced nature of business process event datasets.


Process prediction Deep learning Feature hashing N-grams Stacked autoencoders 


  1. Barga R, Fontama V, Tok WH, Cabrera-Cordon L (2015) Predictive analytics with Microsoft Azure machine learning. Apress, Berkely, CACrossRefGoogle Scholar
  2. Bergstra JS, Bardenet R, Bengio Y, Kégl B (2011) Algorithms for hyper-parameter optimization. Advances in neural information processing systems. Granada, Spain, pp 2546–2554Google Scholar
  3. Bergstra J, Bengio Y (2012) Random search for hyper-parameter optimization. J Mach Learn Res 13(1):281–305Google Scholar
  4. Bose RPJC, van der Aalst WMP, Žliobaitė I, Pechenizkiy M (2011) Handling concept drift in process mining. In: International conference on advanced information systems engineering, Springer, London, pp 391–405Google Scholar
  5. Bradley AP (1997) The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognit 30(7):1145–1159CrossRefGoogle Scholar
  6. Breuker D, Matzner M, Delfmann P, Becker J (2016) Comprehensible predictive models for business processes. MIS Q 40(4):1009–1034CrossRefGoogle Scholar
  7. Candel A, Parmar V, LeDell E, Arora A (2016) Deep learning with h2o. H2O Inc, CAGoogle Scholar
  8. Caragea C, Silvescu A, Mitra P (2012) Protein sequence classification using feature hashing. Proteome Sci 10(1):1–14CrossRefGoogle Scholar
  9. Caruana R, Karampatziakis N, Yessenalina A (2008) An empirical evaluation of supervised learning in high dimensions. In: 25th international conference on machine learning, ACM, Helsinki, pp 96–103Google Scholar
  10. Caruana R, Niculescu-Mizil A (2006) An empirical comparison of supervised learning algorithms. In: 23rd international conference on machine learning. ACM, Pittsburgh, pp 161–168Google Scholar
  11. Da Silva NFF, Hruschka ER, Hruschka ER (2014) Tweet sentiment analysis with classifier ensembles. Decis Support Syst 66:170–179CrossRefGoogle Scholar
  12. Davenport TH, Harris JG (2007) Competing on analytics: the new science of winning. Harvard Business School Press, BostonGoogle Scholar
  13. Di Francescomarino C, Dumas M, Federici M, et al (2016) Predictive business process monitoring framework with hyperparameter optimization. In: 28th international conference on advanced information systems engineering, Springer, Ljubljana, pp 361–376Google Scholar
  14. Duan L, Da Xu L (2012) Business intelligence for enterprise systems: a survey. IEEE Trans Ind Inform 8(3):679–687CrossRefGoogle Scholar
  15. Erhan D, Bengio Y, Courville A et al (2010) Why does unsupervised pre-training help deep learning? J Mach Learn Res 11:625–660Google Scholar
  16. Evermann J, Rehse J-R, Fettke P (2017) Predicting process behaviour using deep learning. Decis Support Syst 100:129–140CrossRefGoogle Scholar
  17. Folino F, Guarascio M, Pontieri L (2012) Discovering context-aware models for predicting business process performances. In: OTM confederated international conferences “on the move to meaningful internet systems”, Springer, Rome, pp 287–304Google Scholar
  18. Forman G, Kirshenbaum E (2008) Extremely fast text feature extraction for classification and indexing. In: 17th ACM conference on information and knowledge management. ACM, Napa Valley, pp 1221–1230Google Scholar
  19. Ganchev K, Dredze M (2008) Small statistical models by random feature mixing. In: the ACL08 HLT workshop on mobile language processing, Columbus, OH, pp 19–20Google Scholar
  20. Goodfellow IJ, Warde-Farley D, Mirza M, et al (2013) Maxout networks. (preprint) arXiv arXiv:1302.4389. Accessed 30 Oct 2017
  21. Gregor S, Hevner AR (2013) Positioning and presenting design science research for maximum impact. MIS Q 37(2):337–356CrossRefGoogle Scholar
  22. Hall M, Frank E, Holmes G et al (2009) The WEKA data mining software: an update. ACM SIGKDD Explor Newslett 11(1):10–18CrossRefGoogle Scholar
  23. Han H, Wang W-Y, Mao B-H (2005) Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning. In: Advances in Intelligent Computing. ICIC 2005. Lecture Notes in Computer Science, vol 3644. Springer, Berlin, Heidelberg, pp 878–887Google Scholar
  24. Hinton GE, Osindero S, Teh Y-W (2006) A fast learning algorithm for deep belief nets. Neural Comput 18(7):1527–1554CrossRefGoogle Scholar
  25. Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science 313(5786):504–507CrossRefGoogle Scholar
  26. Huang C, Li Y, Change Loy C, Tang X (2016) Learning deep representation for imbalanced classification. In: IEEE conference on computer vision and pattern recognition, IEEE, Las Vegas, pp 5375–5384Google Scholar
  27. Izadyyazdanabadi M, Belykh E, Mooney M et al (2017) Convolutional neural networks: ensemble modeling, fine-tuning and unsupervised semantic localization for intraoperative CLE images. Accessed 30 Oct 2017
  28. Kang B, Kim D, Kang S (2012a) Periodic performance prediction for real-time business process monitoring. Ind Manag Data Syst 112(1):4–23CrossRefGoogle Scholar
  29. Kang B, Kim D, Kang S-H (2012b) Real-time business process monitoring method for prediction of abnormal termination using KNNI-based LOF prediction. Expert Syst Appl 39(5):6061–6068CrossRefGoogle Scholar
  30. Lakshmanan GT, Shamsi D, Doganata YN et al (2015) A markov prediction model for data-driven semi-structured business processes. Knowl Inf Syst 42(1):97–126CrossRefGoogle Scholar
  31. Langford J, Li L, Strehl A (2007) Vowpal wabbit online learning project. Technical report, Yahoo!Google Scholar
  32. Larochelle H, Erhan D, Courville A et al (2007) An empirical evaluation of deep architectures on problems with many factors of variation. In: 24th international conference on machine larning, ACM, Corvallis, pp 473–480Google Scholar
  33. LaValle S, Lesser E, Shockley R et al (2011) Big data, analytics and the path from insights to value. MIT Sloan Manag Rev 52:21–32Google Scholar
  34. Le M, Gabrys B, Nauck D (2017) A hybrid model for business process event and outcome prediction. Expert Syst 34(5):e12079CrossRefGoogle Scholar
  35. Le M, Nauck D, Gabrys B, Martin T (2014) Sequential clustering for event sequences and its impact on next process step prediction. In: International conference on information processing and management of uncertainty in knowledge-based systems, Springer, Cádiz, pp 168–178Google Scholar
  36. LeCun YA, Bottou L, Orr GB, Müller KR (2012) Efficient backprop. Neural networks: tricks of the trade. Springer, Berlin, pp 9–50CrossRefGoogle Scholar
  37. Leontjeva A, Conforti R, Di Francescomarino C, et al (2015) Complex symbolic sequence encodings for predictive monitoring of business processes. In: International conference on business process management, Springer, Innsbruck, pp 297–313Google Scholar
  38. Márquez-Chamorro AE, Resinas M, Ruiz-Cortés A, Toro M (2017) Run-time prediction of business process indicators using evolutionary decision rules. Expert Syst Appl 87:1–14CrossRefGoogle Scholar
  39. Mehdiyev N, Evermann J, Fettke P (2017) A multi-stage deep learning approach for business process event prediction. In: IEEE 19th conference on business informatics, IEEE, Thessaloniki, pp 119–128Google Scholar
  40. Metzger A, Leitner P, Ivanovic D et al (2015) Comparing and combining predictive business process monitoring techniques. IEEE Trans Syst, Man, Cybern Syst 45(2):276–290CrossRefGoogle Scholar
  41. Polato M, Sperduti A, Burattin A, de Leoni M (2016) Time and activity sequence prediction of business process instances. Accessed 01 Sept 2017
  42. Robnik-Šikonja M (2014) Data generator based on RBF network. (preprint) arXiv arXiv:1403.7308. Accessed 01 Sept 2017
  43. Rogge-Solti A, Weske M (2013) Prediction of remaining service execution time using stochastic petri nets with arbitrary firing delays. In: International conference on service-oriented computing, Springer, Berlin, pp 389–403Google Scholar
  44. Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117CrossRefGoogle Scholar
  45. Senderovich A, Di Francescomarino C, Ghidini C et al (2017) Intra and inter-case features in predictive process monitoring: a tale of two dimensions. In: International conference on business process management, Springer, Barcelona, pp 306–323Google Scholar
  46. Shi S, Chu X (2017) Speeding up convolutional neural networks by exploiting the sparsity of rectifier units. Accessed 30 Oct 2017
  47. Steeman W (2013) BPI challenge 2013. Accessed 01 Sept 2017
  48. Sun Z, Pambel F, Wang F (2015) Incorporating big data analytics into enterprise information systems. In: Information and communication technology: third IFIP TC 5/8 international conference, ICT-EurAsia 2015, and 9th IFIP WG 8.9 working conference, CONFENIS 2015, Springer, Daejeon, pp 300–309Google Scholar
  49. Sun Y, Wong AKC, Kamel MS (2009) Classification of imbalanced data: a review. Int J Pattern Recognit Artif Intell 23(4):687–719CrossRefGoogle Scholar
  50. Tax N, Verenich I, La Rosa M, Dumas M (2017) Predictive business process monitoring with LSTM neural networks. In: International conference on advanced information systems engineering, Springer, Essen, pp 477–492Google Scholar
  51. Tomović A, Janičić P, Kešelj V (2006) n-Gram-based classification and unsupervised hierarchical clustering of genome sequences. Comput Methods Programs Biomed 81(2):137–153CrossRefGoogle Scholar
  52. Tu JV (1996) Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. J Clin Epidemiol 49(11):1225–1231CrossRefGoogle Scholar
  53. Unuvar M, Lakshmanan GT, Doganata YN (2016) Leveraging path information to generate predictions for parallel business processes. Knowl Inf Syst 47(2):433–461CrossRefGoogle Scholar
  54. van der Aalst WMP, Schonenberg MH, Song M (2011) Time prediction based on process mining. Inf Syst 36(2):450–475CrossRefGoogle Scholar
  55. van Dongen BF (2012) BPI challenge 2012. Accessed 01 Sept 2017
  56. van Dongen BF, Crooy RA, van der Aalst WMP (2008) Cycle time prediction: when will this case finally be finished? In: OTM confederated international conferences “on the move to meaningful internet systems”, Springer, Monterey, pp 319–336Google Scholar
  57. Verenich I (2016) Helpdesk. Accessed 01 Sept 2017
  58. Vincent P, Larochelle H, Bengio Y, Manzagol P-A (2008) Extracting and composing robust features with denoising autoencoders. In: 25th international conference on machine learning, ACM, Helsinki, pp 1096–1103Google Scholar
  59. Vincent P, Larochelle H, Lajoie I et al (2010) Stacked denoising autoencoders: learning useful representations in a deep network with a local denoising criterion. J Mach Learn Res 11:3371–3408Google Scholar
  60. Wang S, Yao X (2012) Multiclass imbalance problems: analysis and potential solutions. IEEE Trans Syst Man Cybern Part B (Cybernetics) 42(4):1119–1130CrossRefGoogle Scholar
  61. Weinberger K, Dasgupta A, Langford J, et al (2009) Feature hashing for large scale multitask learning. In: Proceedings of the 26th annual international conference on machine learning – ICML’09, ACM, Montreal, pp 1–8Google Scholar
  62. Wickham H, Francois R (2015) dplyr: a grammar of data manipulation. R Package Version 04(1):20Google Scholar
  63. Witt N, Seifert C (2017) Understanding the influence of hyperparameters on text embeddings for text classification tasks. In: International conference on theory and practice of digital libraries, Springer, Thessaloniki, pp 193–204Google Scholar
  64. Wu X, Kumar V, Quinlan JR et al (2008) Top 10 algorithms in data mining. Knowl Inf Syst 14(1):1–37CrossRefGoogle Scholar
  65. XES Standard (2016) 1849–2016-IEEE standard for eXtensible event stream (XES) for achieving interoperability in event logs and event streams. Accessed 01 Sept 2017

Copyright information

© Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2018

Authors and Affiliations

  • Nijat Mehdiyev
    • 1
    • 2
    Email author
  • Joerg Evermann
    • 3
  • Peter Fettke
    • 1
    • 2
  1. 1.Institute for Information Systems (IWi), German Research Center for Artificial Intelligence (DFKI)SaarbrueckenGermany
  2. 2.Saarland UniversitySaarbrueckenGermany
  3. 3.Memorial University of NewfoundlandSt. John’sCanada

Personalised recommendations