Download Visualisation and Process Mining

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts
no text concepts found
Transcript
Visualization in Process Mining
Proposed Alternative Titles
Survey of graph visualization techniques in process mining
Visualizing process models: requirements, analysis, system
evaluation
Proposed aim and Content of the paper
 Aim:
Carry out a survey of the visualization techniques that are
currently being used to present process mining results and to
Evaluate their effectiveness.
Optional Addition:
Come up with an approach that will endeavour overcome the
limitations that have been identified from the evaluation
(possibly a combination of graph visualizations)
Proposed Structure (1)
The paper will have 5 sections, that is;
 Introduction
 Key concepts and definitions
 Review of visualization techniques used in process mining
(Visualization techniques)
 Evaluation of visualization Techniques
 Conclusion
Proposed Structure (2)
The paper will have 6 sections, that is;
 Introduction
 Key concepts and definitions
 Review of visualization techniques used in process mining
(Visualization techniques)
 Evaluation of visualization Techniques
 Proposed approach
 Conclusion
Introduction
 This section will give an overview of the process mining
outlining the challenges.
 It will also give an overview of why visualization can be
used as mean to solving some of the challenges i.e.
Displaying complex or large process models arising from
process mining.
Key concepts and definitions
 Process Mining
 What it is.
 Aim of process mining.
 Visualization
 What it is.
 Purpose.
Visualization Techniques
 Graph Visualization
 This is a technique that is used to display data that has elements
that have a relation between them. Data presentation is
characterized by nodes and edges; the nodes representing the
data elements and the edges represent the relation between
them.
 Graph visualization has been applied in various domains
including real-time systems represented as state-transition
system
Graph visualization cont’d
 In process mining, process models may be constructed as
Petri nets or EPCs. As Petri nets, the data elements can take
either of two kinds of nodes; a place or a transition and the
relation between them is represented as directed edges.
 Each graph has a layout associated to it. This determines the
way the nodes and edges appear or are displayed.
 The most commonly used graph layout in process mining is
the tree layout.
Matrix Visualization
 Matrices
 This involves the presentation of data inform of a matrix (rows
and columns in a matrix) and converting this data matrix to a
matrix map. Each data entry is represented as by a dot or square
of a given colour (Legát, 2005).
 This has been used in the dotted chat and the cloud chamber.
Some of the performance analysis results are displayed using this
technique.
Clustering
 Clustering
 Clustering is a process of grouping data or discovering classes of
data based on a number of conditions or rules.
 Clustering is a concept used in visualization to group related
data, as a means to reduce on complexity and space being
utilized to display a large graph (Herman et al., 2000).
 This technique has not been used independently in process
mining results. It has been used to clusters related tasks or
events in EPC process model diagrams for example it is has
been used in conjunction with the fuzzy miner.