Anomaly Detection based on Control-flow Pattern of Parallel Business Processes

[...]Alpha miner algorithms as the above-mentioned processed event logs directly without data filtering, anomaly business process due to the existence of low-frequency values on several trace clusters. [...]this paper proposed a method to tackle this problem by using a graph database incorporating t...

Full description

Saved in:
Bibliographic Details
Published in:Telkomnika 2018-12, Vol.16 (6), p.2809-2816
Main Authors: Darmawan, Hendra, Sarno, Riyanarto, Ahmadiyah, Adhatus Solichah, Sungkono, Kelly Rossa, Wahyuni, Cahyaningtyas Sekar
Format: Article
Language:English
Subjects:
Algorithms
Anomalies
Business process engineering
Clustering
Filtration
Flow distribution
Interlibrary loans
Names
Queries
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:[...]Alpha miner algorithms as the above-mentioned processed event logs directly without data filtering, anomaly business process due to the existence of low-frequency values on several trace clusters. [...]this paper proposed a method to tackle this problem by using a graph database incorporating trace clustering and data filtering. The objective of this step is to remove the anomaly data. Because the Neo4J cannot save the result of a query, automatic deletion cannot be done using Neo4J. Table 1 describes the content of event logs of interlibrary loan application used in this research. In this paper, the number of NEXT relation between activity Request material and Check availability activity is 70 relations. [...]it is necessary to delete NEXT relation and then redo sequence discovery to form the correct business process model.
ISSN:1693-6930
2302-9293
DOI:10.12928/TELKOMNIKA.V16Í6.10568