Process discovery methods automatically infer process models based on events logs that are recorded by information systems. Several heuristic process discovery methods have been proposed to cope with less structured processes and the presence of noise in the event log. However, (1) a large parameter space needs to be explored, (2) several of the many available heuristics can be chosen from, (3) data attributes are not used for discovery, (4) discovered models are not visualized as described in literature, and (5) existing tools do not give reliable quality diagnostics for discovered models. We present the interactive Data-aware Heuristics Miner (iDHM), a modular tool that attempts to address those five issues. The iDHM enables quick interactive exploration of the parameter space and several heuristics. It uses data attributes to improve the discovery procedure and provides built-in conformance checking to get direct feedback on the quality of the model. It is the first tool that visualizes models using the concise Causal Net (C-Net) notation. We provide a walk-through of the iDHM by applying it to a large event log with hospital billing information.
[1]
Wil M.P. van der Aalst,et al.
Fuzzy Mining - Adaptive Process Simplification Based on Multi-perspective Metrics
,
2007,
BPM.
[2]
Hajo A. Reijers,et al.
Decision Mining Revisited - Discovering Overlapping Rules
,
2016,
CAiSE.
[3]
Seppe K. L. M. vanden Broucke,et al.
Fodina: A robust and flexible heuristic process discovery technique
,
2017,
Decis. Support Syst..
[4]
Hajo A. Reijers,et al.
Balanced multi-perspective checking of process conformance
,
2016,
Computing.
[5]
Wil M. P. van der Aalst,et al.
Process Mining
,
2016,
Springer Berlin Heidelberg.
[6]
Marlon Dumas,et al.
Automated Discovery of Structured Process Models: Discover Structured vs. Discover and Structure
,
2016,
ER.
[7]
A. J. M. M. Weijters,et al.
Flexible Heuristics Miner (FHM)
,
2011,
2011 IEEE Symposium on Computational Intelligence and Data Mining (CIDM).
[8]
Hajo A. Reijers,et al.
Data-driven process discovery
,
2017
.