VaryMinions: Leveraging RNNs to Identify Variants in Event Logs

Sophie Fortz; Paul Temple; Xavier DEVROEY; Patrick HEYMANS; GILLES PERROUIN

VaryMinions: Leveraging RNNs to Identify Variants in Event Logs

Sophie Fortz, Paul Temple, Xavier DEVROEY, Patrick HEYMANS, GILLES PERROUIN

Research output: Contribution in Book/Catalog/Report/Conference proceeding › Conference contribution

132 Downloads (Pure)

Abstract

Business processes have to manage variability in their execution, eg to deliver the correct building permit in different municipalities. This variability is visible in event logs, where sequences of events are shared by the core process (building permit authorisation) but may also be specific to each municipality. To rationalise resources (eg derive a configurable business process capturing all municipalities' permit variants) or to debug anomalous behaviours, it is mandatory to identify to which variant a given trace belongs. This paper supports this task by training Long Short Term Memory (LSTMs) and Gated Recurrent Units (GRUs) algorithms on two datasets: a configurable municipality and a travel expenses workflow. We demonstrate that variability can be identified accurately (>87%) and discuss the challenges of learning highly entangled variants.

Original language	English
Title of host publication	5th International Workshop on Machine Learning Techniques for Software Quality Evolution
Publisher	ACM Press
Publication status	Published - 2021

Access to Document

VaryMinions_MALTESQUE2021Accepted author manuscript, 447 KB

LIFTS: Learning Featured Transition Systems
Fortz, S., Perrouin, G. & Heymans, P.
1/10/20 → 31/12/21
Project: Research
VeriLearn: Verifying Learning Artificial Intelligence Systems
Heymans, P., Frénay, B., Schobbens, P. Y., Temple, P., Nanfack, G., Amrani, M. & Bibal, A.
1/01/18 → 31/12/22
Project: Research

Cite this

@inproceedings{d2626b47ae36477e816abcc3e05a8edd,

title = "VaryMinions: Leveraging RNNs to Identify Variants in Event Logs",

abstract = "Business processes have to manage variability in their execution, eg to deliver the correct building permit in different municipalities. This variability is visible in event logs, where sequences of events are shared by the core process (building permit authorisation) but may also be specific to each municipality. To rationalise resources (eg derive a configurable business process capturing all municipalities' permit variants) or to debug anomalous behaviours, it is mandatory to identify to which variant a given trace belongs. This paper supports this task by training Long Short Term Memory (LSTMs) and Gated Recurrent Units (GRUs) algorithms on two datasets: a configurable municipality and a travel expenses workflow. We demonstrate that variability can be identified accurately (>87%) and discuss the challenges of learning highly entangled variants.",

author = "Sophie Fortz and Paul Temple and Xavier DEVROEY and Patrick HEYMANS and GILLES PERROUIN",

year = "2021",

language = "English",

booktitle = "5th International Workshop on Machine Learning Techniques for Software Quality Evolution",

publisher = "ACM Press",

address = "United States",

}

TY - GEN

T1 - VaryMinions

T2 - Leveraging RNNs to Identify Variants in Event Logs

AU - Fortz, Sophie

AU - Temple, Paul

AU - DEVROEY, Xavier

AU - HEYMANS, Patrick

AU - PERROUIN, GILLES

PY - 2021

Y1 - 2021

N2 - Business processes have to manage variability in their execution, eg to deliver the correct building permit in different municipalities. This variability is visible in event logs, where sequences of events are shared by the core process (building permit authorisation) but may also be specific to each municipality. To rationalise resources (eg derive a configurable business process capturing all municipalities' permit variants) or to debug anomalous behaviours, it is mandatory to identify to which variant a given trace belongs. This paper supports this task by training Long Short Term Memory (LSTMs) and Gated Recurrent Units (GRUs) algorithms on two datasets: a configurable municipality and a travel expenses workflow. We demonstrate that variability can be identified accurately (>87%) and discuss the challenges of learning highly entangled variants.

AB - Business processes have to manage variability in their execution, eg to deliver the correct building permit in different municipalities. This variability is visible in event logs, where sequences of events are shared by the core process (building permit authorisation) but may also be specific to each municipality. To rationalise resources (eg derive a configurable business process capturing all municipalities' permit variants) or to debug anomalous behaviours, it is mandatory to identify to which variant a given trace belongs. This paper supports this task by training Long Short Term Memory (LSTMs) and Gated Recurrent Units (GRUs) algorithms on two datasets: a configurable municipality and a travel expenses workflow. We demonstrate that variability can be identified accurately (>87%) and discuss the challenges of learning highly entangled variants.

M3 - Conference contribution

BT - 5th International Workshop on Machine Learning Techniques for Software Quality Evolution

PB - ACM Press

ER -

VaryMinions: Leveraging RNNs to Identify Variants in Event Logs

Abstract

Access to Document

Fingerprint

Projects

LIFTS: Learning Featured Transition Systems

VeriLearn: Verifying Learning Artificial Intelligence Systems

Cite this