Towards Knowledge-Centric Process Mining

TL;DR

This paper introduces a knowledge graph-based approach to improve process mining in noisy or incomplete event logs, enhancing process analysis and understanding of variability.

Contribution

It presents a novel method that uses knowledge graphs to handle noise and incompleteness in event logs for process mining.

Findings

Improves process analysis accuracy in noisy logs

Supports understanding of process variability

Enhances robustness of process mining techniques

Abstract

Process analytic approaches play a critical role in supporting the practice of business process management and continuous process improvement by leveraging process-related data to identify performance bottlenecks, extracting insights about reducing costs and optimizing the utilization of available resources. Process analytic techniques often have to contend with real-world settings where available logs are noisy or incomplete. In this paper we present an approach that permits process analytics techniques to deliver value in the face of noisy/incomplete event logs. Our approach leverages knowledge graphs to mitigate the effects of noise in event logs while supporting process analysts in understanding variability associated with event logs.