Skip to Content
You are currently on the new version of our website. Access the old version .
AlgorithmsAlgorithms
  • Article
  • Open Access

25 May 2022

21 Pages

Data Preprocessing Method and API for Mining Processes from Cloud-Based Application Event Logs

and
School of Electrical Engineering and Computer Science, University of Ottawa, Ottawa, ON K1N 6N5, Canada
*
Author to whom correspondence should be addressed.
This article belongs to the Special Issue Process Mining and Its Applications

Abstract

Process mining (PM) exploits event logs to obtain meaningful information about the processes that produced them. As the number of applications developed on cloud infrastructures is increasing, it becomes important to study and discover their underlying processes. However, many current PM technologies face challenges in dealing with complex and large event logs from cloud applications, especially when they have little structure (e.g., clickstreams). By using Design Science Research, this paper introduces a new method, called cloud pattern API-process mining (CPA-PM), which enables the discovery and analysis of cloud-based application processes using PM in a way that addresses many of these challenges. CPA-PM exploits a new application programming interface, with an R implementation, for creating repeatable scripts that preprocess event logs collected from such applications. Applying CPA-PM to a case with real and evolving event logs related to the trial process of a software-as-a-service cloud application led to useful analyses and insights, with reusable scripts. CPA-PM helps producing executable scripts for filtering event logs from clickstream and cloud-based applications, where the scripts can be used in pipelines while minimizing the need for error-prone and time-consuming manual filtering.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.