Book chapter
Duration-Aware Alignment of Process Traces
Advances in Data Mining. Applications and Theoretical Aspects, pp 379-393
28 Jun 2016
Abstract
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
Details
- Title
- Duration-Aware Alignment of Process Traces
- Creators
- Sen Yang - Department of Electrical and Computer Engineering, Rutgers University, Piscataway, USAMoliang Zhou - Department of Electrical and Computer Engineering, Rutgers University, Piscataway, USARachel Webman - Division of Trauma and Burn Surgery, Children’s Nat’l Medical Center, Washington, D.C., USAJaeWon Yang - Division of Trauma and Burn Surgery, Children’s Nat’l Medical Center, Washington, D.C., USAAleksandra Sarcevic - College of Information Science and Technology, Drexel University, Philadelphia, USAIvan Marsic - Department of Electrical and Computer Engineering, Rutgers University, Piscataway, USARandall S Burd - Division of Trauma and Burn Surgery, Children’s Nat’l Medical Center, Washington, D.C., USA
- Publication Details
- Advances in Data Mining. Applications and Theoretical Aspects, pp 379-393
- Series
- Lecture Notes in Computer Science
- Publisher
- Springer International Publishing; Cham
- Resource Type
- Book chapter
- Language
- English
- Academic Unit
- Information Science
- Web of Science ID
- WOS:000389647400028
- Scopus ID
- 2-s2.0-84978937943
- Other Identifier
- 991014976892804721
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