Extensions, Analysis and Experimental Assessment of a Probabilistic Ensemble-learning Framework for Detecting Deviances in Business Process Instances

Alfredo Cuzzocrea, Francesco Folino, Massimo Guarascio, Luigi Pontieri

2017

Abstract

This paper significantly extends a previous proposal where an innovative ensemble-learning framework for mining business process deviances that exploits multi-view learning has been provided. Here, we introduce some relevant contributions: (i) a further learning method that extends and refines the previous methods via introducing the idea of probabilistically combining different deviance detection models (DDMs); (ii) a complete conceptual architecture that implements the extended multi-view ensemble-learning framework; (iii) a wide and comprehensive experimental assessment of the framework, even in comparison with existent competitors. The investigated scientific context falls in the so-called Business Process Intelligence (BPI) research area, which is relevant for a wide number of real-life applications. These novel contributions clearly confirm the flexibility, the reliability and the effectiveness of the general deviance detection framework, respectively.

Paper Citation

in Harvard Style

Cuzzocrea A., Folino F., Guarascio M. and Pontieri L. (2017). Extensions, Analysis and Experimental Assessment of a Probabilistic Ensemble-learning Framework for Detecting Deviances in Business Process Instances . In Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-247-9, pages 162-173. DOI: 10.5220/0006340001620173

in Bibtex Style

@conference{iceis17,
author={Alfredo Cuzzocrea and Francesco Folino and Massimo Guarascio and Luigi Pontieri},
title={Extensions, Analysis and Experimental Assessment of a Probabilistic Ensemble-learning Framework for Detecting Deviances in Business Process Instances},
booktitle={Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2017},
pages={162-173},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006340001620173},
isbn={978-989-758-247-9},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 19th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - Extensions, Analysis and Experimental Assessment of a Probabilistic Ensemble-learning Framework for Detecting Deviances in Business Process Instances
SN - 978-989-758-247-9
AU - Cuzzocrea A.
AU - Folino F.
AU - Guarascio M.
AU - Pontieri L.
PY - 2017
SP - 162
EP - 173
DO - 10.5220/0006340001620173