Generating LTL Formula for Verification of Business Process Event Logs Using Decision Tree Learning

Hiroki HORITA  Hideaki HIRAYAMA  Takeo HAYASE  Yasuyuki TAHARA  Akihiko OHSUGA  

D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)   Vol.J101-D   No.3   pp.530-538
Publication Date: 2018/03/01
Online ISSN: 1881-0225
Type of Manuscript: Special Section PAPER (Special Section on Student Research)
business process,  verification,  machine learning,  process mining,  

Full Text(in Japanese): PDF(1.2MB)
>>Buy this Article

Process mining is a important means for analyzing business process and LTL checker is a famous tool for process mining. However, since many business analysts are not familiar with mathematical notation like LTL, it is difficult to describe exactly the property to be verified when describing the property to be verified in the business process is there. Therefore, in this study, learning is performed based on feature quantities extracted from the business process execution log using a decision tree, and a logical expression is automatically generated. We propose a method to describe properties to be verified.