PV226 – Lasaris seminar / 2021 / R. Ošlejšek 1
Data Visualization – Principles and Challenges
Radek Ošlejšek
Data Visualization
●
Avoiding confusion: IT IS NOT about the design of GUI, e.g., where to place info
window, what color to choose (although these UX aspects are part of any good
graphical tool).
●
Three main fields in visualization
– Scientific visualization (SciVis)
– Information visualization (InfoVis)
– Visual analytics (VA) [Furmanová et al.: Multiscale Visual Drilldown for the Analysis
of Large Ensembles of Multi-Body Protein Complexes, 2019]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 3
Scientific visualization (SciVis)
●
Goal: To graphically illustrate scientific phenomena
– Techniques to create a renderable geometric model of the data
– Explanatory visualizations: Graphics representation is used for understanding and
interpretation
– Studied phenomenon is usually modeled by measurements
●
Interdisciplinary: medical, molecular, flow, volumetric … visualization
A visual approach to illustrate the complex relationships within a Supernova
[Ma et al.: Scientific discovery through advanced visualization, 2005]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 4
SciVis: Face Morpohology
●
Research activities:
– Similarity of Human Faces in to cooperation with doc. Petra Urbanová,
Department of Anthropology, Faculty of Science, MU.
– FIDENTIS Analyst 2: https://www.radek-oslejsek.cz/it/fidentis-analyst-2/
●
For anthropologists, police, (plastic) surgeons
●
Exercise: Given two 3D scans of human faces, what visualization techniques we
can use to compare their (dis)similarity?
[Ferková, Z.: Comparison and Analysis of Multiple 3D Shapes, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 5
SciVis: Similarity of Human Faces
●
Approach: Overlapping with transparency
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 6
SciVis: Similarity of Human Faces
●
Approach: Overlapping with contours
[Furmanová. K.: Visualization techniques for 3D facial comparison, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 7
SciVis: Similarity of Human Faces
●
Approach: Overlapping with gliphs
[Furmanová. K.: Visualization techniques for 3D facial comparison, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 8
SciVis: Similarity of Human Faces
●
Approach: Overlapping with volume highlighting (by fog)
[Furmanová. K.: Visualization techniques for 3D facial comparison, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 9
SciVis: Similarity of Human Faces
●
Approach: Distance heatmap of Hausdorff distance
[Furmanová. K.: Visualization techniques for 3D facial comparison, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 10
Information visualization (InfoVis)
●
Goal: To provide (interactive) visual representations of abstract data to reinforce
human cognition
– Main focus on representing data in easily understandable way, supported by
intuitive interaction
– Transformation of data and phenomenons to complementary, often 2D, views
– Distribution of the data in space or time, relationships between data values, etc.
●
The most common uses:
– Presentation, exploratory analysis, confirmation analysis
[Keim et al.: FinDEx: A spectral visualization system for analyzing financial time series data., 2006]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 11
SciVis vs. InfoVis
●
Explanatory (SciVis) vs. exploratory or confirmation analysis (InfoVis)
●
Direct visualization of the phenomena (SciVis) vs. indirect views of phenomena aspects
(e.g., the distribution of the data in space or time)
●
It is SciVis when the spatial representation is given, and InfoVis when the spatial
representation is chosen
[Hamminger, B.: Information Visualization, https://slideplayer.com/slide/4644055/]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 12
SciVis vs. InfoVis – Face Morphology Example
●
SciVis: visualize distances in a 3D heatmap
●
InfoVis: View of data distribution (histogram), filtering (e.g., the noise in the eyebrow areas),
mutual similarity of N faces, etc.
[Furmanová. K.: Visualization techniques for 3D facial comparison, 2015]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 13
InfoVis Techniques
●
Many visualization techniques has been designed for concrete data characteristics
– Spatial data, geospatial data, multivariate data, trees and networks, …
●
Many visualization techniques has been designed for concrete goals
– Data distribution, comparison, composition, etc.
●
Still, the design of concrete solution is challenging
– www.datavizproject.com
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 14
Visual Analytics (VA)
●
Goal: Analytical reasoning supported by interactive visual interfaces
– Hypotheses-based (what-if) analyses
– Revealing relationships hidden in the data
– Building the knowledge (step by step) from the data
●
Analytical goals of domain experts (users) drive the VA solution
[Keim et al.: Visual Analytics: Scope and Challenges, 2008][Keim et al.: Visual Analytics, 2009]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 15
VA – Face Morphology Example
●
High-level analytical goals (examples):
– Face symmetry: Modeling for plastic surgery
– Morphing: Estimation how a face will change when a child grows up
– Identification: Searching for a similar faces in a database, e.g., by police
– Classification: Identification of clusters in a set of cases, e.g., by race, gender, age,
etc.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 16
VA methodology
●
The development of a really useful VA tool is challenging. It is necessary to follow
many rules and best practices to achieve good results and to prove usability
– Tight cooperation with domain experts for both requirements analysis and usability
evaluation
– Using iterative design methodologies, e.g., user-centered design (it isn’t an ad-hoc
process)
– Formal evaluation of results, e.g., quantitative and qualitative methods of measuring
user experience
●
The development process can be considered a special discipline of software
engineering
UNDERSTAND EXPLORE MATERIALIZE
Visual Analytics for
Practical Cybersecurity Training
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 18
Problem statement
No tangible output (like a code in programming courses)
●
Tutors have no idea, what trainees do, whether they are stacked in some task, etc.
●
Trainees don’t know whether what they did wrong, or whether there was a faster
solution to the tasks.
●
Training designers don’t know whether the game was too easy or difficult.
●
Research Goal: To research and develop data analysis tools providing insight into
educational aspects of cybersecurity training and enable comparison, assessment, and
continuous improvement.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 19
VA for Practical Cybersecurity Training: Data
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 20
KYPO Cyber Range Platform
Remote connection
Collected data:
●
Commands executed on hosts
●
Exercise milestones, e.g.,
– Taking hints
– (Un)successfully solved tasks
●
Assessment (penalties)
Format:
●
Event logs
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 21
VA for Practical Cybersecurity Training: Knowledge
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 22
VA for Practical Cybersecurity Training: Knowledge
OŠLEJŠEK, Radek, Vít RUSŇÁK, Karolína DOČKALOVÁ BURSKÁ, Valdemar ŠVÁBENSKÝ, Jan VYKOPAL and Jakub ČEGAN.
Conceptual Model of Visual Analytics for Hands-on Cybersecurity Training.
In IEEE Transactions on Visualization and Computer Graphics, 2021.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 23
VA for Practical Cybersecurity Training: Models
VA for Practical Cybersecurity Training: Models
●
Statistical models
●
Process mining models
– Challenge: Data pre-processing and mapping affect obtained graphs
– Challenge: The selection of process discovery algorithm affects obtained graphs
– Challenge: Problem with the scalability of obtained graphs
●
What is the same or sufficiently similar commands?
– User 1: ssh root@147.251.8.28
– User 2: ssh 147.251.8.28
– User 3: ssh -4 root@147.251.8.28
– User 4: ssh 127.1.5.8
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 25
Data aggregation and filtering
LevelStarted (4)
ssh (4)
4
... (4)
4
LevelStarted (4)
ssh 147.251.8.28 (3)
3
... (3)
3
ssh 127.1.5.8 (1)
... (3)
1
1
LevelStarted (4)
ssh root@147.251.8.28 (2)
2
... (3)
3
ssh 127.1.5.8 (1)
... (3)
1
1
ssh 147.251.8.28 (1)
1
... (3)
1
MACÁK, Martin, Radek OŠLEJŠEK and Barbora Bühnová. Process Mining Analysis of Puzzle-Based Cybersecurity Training.
Innovation and Technology in Computer Science Education (ITiCSE’22), to appear, 2022.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 26
VA for Practical Cybersecurity Training: Visualizations
●
TO DO
Data-Based Vis: Personalized feedback to trainees
Goal: Learning from own mistakes
●
What did I do wrong in selected tasks?
●
Where I lost most points and why?
●
...
OŠLEJŠEK, Radek, Vít RUSŇÁK, Karolína BURSKÁ, Valdemar ŠVÁBENSKÝ a Jan VYKOPAL.
Visual Feedback for Players of Multi-Level Capture the Flag Games: Field Usability Study.
In IEEE Symposium on Visualization for Cyber Security (VizSec‘19)
VYKOPAL, Jan, Radek OŠLEJŠEK, Karolína BURSKÁ and Kristína ZÁKOPČANOVÁ.
Timely Feedback in Unstructured Cybersecurity Exercises.
In ACM Technical Symposium on Computer Science Education (SIGCSE’18)
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 28
Data-Based Vis: Insight for organizing participants
Goal: Situational awareness and timely intervention
●
Which trainees are in trouble? Why?
●
Is the training session on schedule, or is there some delay?
●
...
DOČKALOVÁ BURSKÁ Karolína, Vít RUSŇÁK and Radek OŠLEJŠEK.
Enhancing Situational Awareness for Tutors of Cybersecurity Capture the Flag Games.
In International Conference Information Visualization (iV’21).
Data-Based Vis: Post-training analysis
Goal: Improve the impact of learning
●
Was training too easy or difficult?
●
What are the sources of losing motivation and giving up the training?
●
Are there some flows in the scenario, requirements, etc.?
●
...
DOČKALOVÁ BURSKÁ Karolína, Vít RUSŇÁK and Radek OŠLEJŠEK.
Data-driven insight into the puzzle-based cybersecurity training.
In Computers & Graphics, 2021.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 30
●
The previous examples demonstrate visualization of data
●
Visualization of process models brings challenges in comprehensibility
and scalability of models.
●
Idea: Provide alternate view to a traditional graph representation
– From the VA perspective, process graphs are so-called
multivariate networks
Model-Based Vis: Process Graphs
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 31
Multivariate Networks
●
No silver bullet solution available.
●
A combination of carefully selected, adapted, and mutually connected
visualizations and interactions is required.
●
A suitable approach depends on the properties of the multivariate network
[Nobre, C. et al.: The state of the art in visualizing multivariate networks. In Computer Graphics Forum, 2019]
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 32
Process Models as Multivariate Networks
●
Properties of our process models:
– Medium (less than 1.000 nodes) or large size (more than 1.000 nodes)
– K-partite
– Heterogeneous nodes with few attributes and homogeneous edges.
●
Suitable visualization tactics:
– Attribute-driven faceting: groups nodes according to one or more
attributes and places the elements of a group in a shared region
– Quilts: A tabular layout optimized for layered networks
– Integrated or juxtaposed view operations.
●
Integrated: the topology and the attribute
visualizations are laid out with the other
view in mind.
●
Juxtaposed: separates the topology
visualization from the attribute visualization
into two or more views.
MACÁK, Martin, Radek OŠLEJŠEK and Barbora Bühnová.
Applying Process Discovery to Cybersecurity Training: An Experience Report.
Cyber Range Technologies and Applications (CACOE’22), under review.
PV226 – Lasaris seminar / 2021 / R. Ošlejšek 33
Infrastructure analysis
Thank you for your attention!