TU Darmstadt / ULB / TUbiblio

ProcessExplorer: An Interactive Visual Recommendation System for Process Mining

Seeliger, Alexander ; Nolle, Timo ; Mühlhäuser, Max (2018)
ProcessExplorer: An Interactive Visual Recommendation System for Process Mining.
Conference or Workshop Item, Bibliographie

Abstract

More and more business process data is collected by organizations to analyze and optimize their process performance. As a consequence it is particularly challenging to locate possible process issues or potential optimizations using process mining. Process mining aims at analyzing the actual usage of information systems by reconstructing a process model from recorded event log. However, such large amount of data often leads to spaghetti-like visualizations which are in-comprehensive and inaccurate. This paper addresses this issue by introducing an unsupervised visual recommender system for process analysis. The system provides suggestions during the interactive visual inspection of the discovered process model by recommending points of interests (e.g., long duration times or exceptional process behavior) ranked by severity. For calculated interest points we characterize the deviation from the average behavior as well as compute the effect the observed conspicuousness has. Our approach has been implemented as a ProM plugin. We evaluate our approach by presenting a case study using a real life event log.

Item Type: Conference or Workshop Item
Erschienen: 2018
Creators: Seeliger, Alexander ; Nolle, Timo ; Mühlhäuser, Max
Type of entry: Bibliographie
Title: ProcessExplorer: An Interactive Visual Recommendation System for Process Mining
Language: English
Date: 20 August 2018
Place of Publication: London, UK
Journal or Publication Title: KDD 2018 Workshop on Interactive Data Exploration and Analytics
Book Title: KDD 2018 Workshop on Interactive Data Exploration and Analytics
URL / URN: http://poloclub.gatech.edu/idea2018/
Abstract:

More and more business process data is collected by organizations to analyze and optimize their process performance. As a consequence it is particularly challenging to locate possible process issues or potential optimizations using process mining. Process mining aims at analyzing the actual usage of information systems by reconstructing a process model from recorded event log. However, such large amount of data often leads to spaghetti-like visualizations which are in-comprehensive and inaccurate. This paper addresses this issue by introducing an unsupervised visual recommender system for process analysis. The system provides suggestions during the interactive visual inspection of the discovered process model by recommending points of interests (e.g., long duration times or exceptional process behavior) ranked by severity. For calculated interest points we characterize the deviation from the average behavior as well as compute the effect the observed conspicuousness has. Our approach has been implemented as a ProM plugin. We evaluate our approach by presenting a case study using a real life event log.

Uncontrolled Keywords: Process mining, Exploratory analysis, Process analysis, Visualization recommendation, Business process intelligence
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Telecooperation
Date Deposited: 19 Jun 2018 09:46
Last Modified: 14 Jun 2021 06:14
PPN:
Export:
Suche nach Titel in: TUfind oder in Google
Send an inquiry Send an inquiry

Options (only for editors)
Show editorial Details Show editorial Details