Selecting Optimal Trace Clustering Pipelines with AutoML

TL;DR

This paper introduces an AutoML framework that recommends optimal trace clustering pipelines for event logs, improving model interpretability and analytics by considering log properties and clustering techniques.

Contribution

The work presents a novel AutoML approach for selecting trace clustering pipelines tailored to specific event logs, addressing a gap in existing clustering methods.

Findings

Framework effectively recommends suitable clustering pipelines.

Experiments with 1000 logs demonstrate improved clustering quality.

Insights into the relationship between log properties and clustering performance.

Abstract

Trace clustering has been extensively used to preprocess event logs. By grouping similar behavior, these techniques guide the identification of sub-logs, producing more understandable models and conformance analytics. Nevertheless, little attention has been posed to the relationship between event log properties and clustering quality. In this work, we propose an Automatic Machine Learning (AutoML) framework to recommend the most suitable pipeline for trace clustering given an event log, which encompasses the encoding method, clustering algorithm, and its hyperparameters. Our experiments were conducted using a thousand event logs, four encoding techniques, and three clustering methods. Results indicate that our framework sheds light on the trace clustering problem and can assist users in choosing the best pipeline considering their scenario.