Exploiting Event Log Event Attributes in RNN Based Prediction

In predictive process analytics, current and historical process data in event logs is used to predict the future, e.g., to predict the next activity or how long a process will still require to complete. Recurrent neural networks (RNN) and its subclasses have been demonstrated to be well suited for creating prediction models. Thus far, event attributes have not been fully utilized in these models. The biggest challenge in exploiting them in prediction models is the potentially large amount of event attributes and attribute values. We present a novel clustering technique that allows for trade-offs between prediction accuracy and the time needed for model training and prediction. As an additional finding, we also find that this clustering method combined with having raw event attribute values in some cases provides even better prediction accuracy at the cost of additional time required for training and prediction.

[1]  C. Furlanello,et al.  Recursive feature elimination with random forest for PTR-MS analysis of agroindustrial products , 2006 .

[2]  Chris H. Q. Ding,et al.  Minimum Redundancy Feature Selection from Microarray Gene Expression Data , 2005, J. Bioinform. Comput. Biol..

[3]  Wil M. P. van der Aalst,et al.  Process Mining - Discovery, Conformance and Enhancement of Business Processes , 2011 .

[4]  Fabrizio Maria Maggi,et al.  Tell me what’s ahead? Predicting remaining activity sequences of business process instances , 2016 .

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Jana-Rebecca Rehse,et al.  Predicting process behaviour using deep learning , 2016, Decis. Support Syst..

[7]  Periklis Andritsos,et al.  Accurate and Transparent Path Prediction Using Process Mining , 2019, ADBIS.

[8]  Fabrizio Maria Maggi,et al.  Predictive Process Monitoring Methods: Which One Suits Me Best? , 2018, BPM.

[9]  Irene Teinemaa,et al.  An interdisciplinary comparison of sequence modeling methods for next-element prediction , 2018, Software and Systems Modeling.

[10]  P. Rousseeuw Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[11]  Markku Hinkka,et al.  Focusing Business Improvements Using Process Mining Based Influence Analysis , 2016, BPM.

[12]  Yoshua Bengio,et al.  On the Properties of Neural Machine Translation: Encoder–Decoder Approaches , 2014, SSST@EMNLP.

[13]  Alessandro Sperduti,et al.  LSTM networks for data-aware remaining time prediction of business process instances , 2017, 2017 IEEE Symposium Series on Computational Intelligence (SSCI).

[14]  Andrew W. Moore,et al.  X-means: Extending K-means with Efficient Estimation of the Number of Clusters , 2000, ICML.

[15]  Stefan Jablonski,et al.  Deep Learning Process Prediction with Discrete and Continuous Data Features , 2018, ENASE.

[16]  Alexander Jung,et al.  Structural Feature Selection for Event Logs , 2017, Business Process Management Workshops.

[17]  Max Mühlhäuser,et al.  BINet: Multivariate Business Process Anomaly Detection Using Deep Learning , 2018, BPM.

[18]  Fabrizio Maria Maggi,et al.  Clustering-Based Predictive Process Monitoring , 2015, IEEE Transactions on Services Computing.

[19]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[20]  Akhil Kumar Business Process Management , 2018, Lecture Notes in Computer Science.

[21]  Ying Zhang,et al.  Batch normalized recurrent neural networks , 2015, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[22]  Alexander Jung,et al.  Classifying Process Instances Using Recurrent Neural Networks , 2018, Business Process Management Workshops.

[23]  Marlon Dumas,et al.  Predictive Business Process Monitoring with LSTM Neural Networks , 2016, CAiSE.