Logo image
Duration-Aware Alignment of Process Traces
Book chapter   Peer reviewed

Duration-Aware Alignment of Process Traces

Sen Yang, Moliang Zhou, Rachel Webman, JaeWon Yang, Aleksandra Sarcevic, Ivan Marsic and Randall S Burd
Advances in Data Mining. Applications and Theoretical Aspects, pp 379-393
28 Jun 2016

Abstract

Reference Alignment Dynamic Time Warping Alignment Algorithm Activity Duration Alignment Accuracy
Objective: To develop an algorithm for aligning process traces that considers activity duration during alignment and helps derive data-driven insights from workflow data. Methods: We developed a duration-aware trace alignment algorithm as part of a Java application that provides visualization of the alignment. The relative weight of the activity type vs. activity duration during the alignment is an adjustable parameter. We evaluated proportional and logarithmic weights for activity duration. Results: We used duration-aware trace alignment on two real-world medical datasets. Compared with existing context-based alignment algorithm, our results show that duration-aware alignment algorithm achieves higher alignment accuracy and provides more intuitive insights for deviation detection and data visualization. Conclusion: Duration-aware trace alignment improves upon an existing trace alignment approach and offers better alignment accuracy and visualization.

Metrics

16 Record Views
8 citations in Scopus

Details

InCites Highlights

Data related to this publication, from InCites Benchmarking & Analytics tool:

Collaboration types
Domestic collaboration
Web of Science research areas
Computer Science, Information Systems
Computer Science, Theory & Methods
Logo image