This document describes Datascape, a 3D immersive data visualization application created by Daden. Datascape allows users to import data from various sources and visualize it in an interactive 3D environment. This overcomes limitations of 2D visualization as it provides additional dimensions to display data attributes and enables easier pattern recognition. Datascape provides tools for data mapping, filtering, grouping and exploring visualized data. Examples shown include line graphs, bar charts, scatter plots and network graphs visualized using Datascape's 3D capabilities.
3. For over 10 years we've had computers with decent
3D graphics, and yet still all we draw is 2D graphs
4. Isn't it about time we leveraged the power of our
computers and started visualising data in 3D?
5. 2D is simple and familiar, but as more and more
data becomes available 2D struggles to cope
6. Adding a 3rd axis lets us easily display another
dimensions worth of data ...
– such as time over a map plot of latitude and
longitude
7. Making the space immersive puts you in there
with your data, you can see it from inside and out
And since the data stays still whilst you move
through it, your brain and eyes find they can relate
to it as they do to the physical world
8. Making it easier to see patterns and anomalies
.. and easier to remember since the brain has
ample visual and spatial cues
9. And we can control not only the 3D position of
each data point, but also its colour, shape, size,
rotation and image
… giving us potentially 8 – 12 dimensions for
every point, sometimes even more
10. To deliver the benefits of immersive 3D data
visualisation Daden have created Datascape
A desktop 3D immersive visual analytics application
LocationtrackingonTwitterGPSdata
11. Datascape sits at the user end of the data analysis
workflow and can work with your existing data
warehouse, ETL and 2D analysis tools
Data Store
Data
Feeds
Live Data Feeds
Extract-Transform-
Load Tools
Datascape
Existing
Analytic Tools
Via
Excel/CSV
or ODBC
12. Datascape has a 5 step process....
1. Import 2. Map 3. Visualise 4. Analyse
Bring data in
by CSV or
ODBC
Assign
fields/columns in
data to properties
of plot point
View and explore
the data, refine
mapping if
necessary
Apply search,
group,
highlight, filter,
animation
techniques etc
5. Re-iterate
13. Data Import
Through the data Source Plug-ins
you can easily import and
manage your data:
• Import data from Excel or
Comma Separated Variables
files
• Use ODBC to bring data
directly in from other
applications and databases,
including live feeds (GPS,
Twitter, RSS), Extract-
Transform-Load tools and Big
Data repositories
• Import network and social
graphs as nodes and edges
files
14. Workspace Management
Workspaces are where
you do your
visualisations and
analysis.
You can create as many
workspaces as you want,
and save them with or
without data.
Tag and filter workspaces
for easy management
You can also copy,
rename and delete
workspaces.
15. Field to Feature Mapping
You can bring multiple datasets
into each workspace.
For each dataset you can define
a unique mapping between the
fields in the data and the plottable
features of each data point:
Position (x,y,z)
Shape
Colour
Size (x,y,z)
Image
Rotation (x,y,z)
Label
URL
Axis Labels (x,y,z)
You can create simple or complex
expressions to link the two, and
we'll publish common choices on
our wiki.
16. Assist and Translation
The Data Assist screen
prompts you with the
available fields, lets you
preview their data, shows
available shapes, colours
and images, and reminds
you of common functions.
Translation Tables let you
map values easily to
colours, shapes and
images with no scripting,
and auto-fill lets the app do
all the initial work to get you
started.
17. Scenery
You can use bars and panels to
create one and two dimensional
axes.
Panels can be assigned images,
including grids, schematics and
maps. You can also assign them
web images via URLs, or
OpenStreetMap.
You can also use a central
sphere, ideal for global plots.
You can also choose between a
variety of images and colours for
the sky and ground.
The navigation tree gives you
easy access to all elements of the
visualisation.
18. Axes and Labels
You'll find that you need a
different sort of axis in an
immersive data plot since you are
not always “outside” the data in a
position to see the axes lines.
Datascape helps to overcome this
in a variety of ways:
• Adding guidelines down to the
XZ plane to help pinpoint
locations on maps
• Adding axis end points on the
horizon along the X, Y and Z
axis
19. Data Interactions
With the data displayed you can
fly through it at will, and look in
any direction from any location.
You can also lock your view to a
single axis, or rotate around a
single point.
You can use the search dialogue
to perform boolean queries on the
data, highlighting selected groups,
and/or manually adding to them.
Groups can be saved, highlighted
and used as the basis for further
searches.
Clicking on any point allows you to
see all that point's data fields, use
them as the basis of a similarity
search, or open a related URL.
20. Groups and Network Lines
Network lines can be
imported alongside node
data, or generated
dynamically to link points by
a chosen field, e.g. all points
for a particular user.
The network lines can have
mappings defined in the
same way as nodes,
controlling colour, shape,
image, size and labels.
You can tab through a group
in order to see each point.
You can also hide the main
data set(s) so as to only
leave one (or more) groups
visible.
21. Dynamic Filtering
You can use the
scrubbing bar to set a
filter window on any field
and manually sweep that
window across the data,
e.g. sweep a 1 week
window across a whole
year's data.
The selected items can
at any time be saved as
a group for later analysis.
22. Live Data
If you are updating the
Datascape database in
real-time from an external
feed/source then you can
update the Datascape
display in real-time too.
Just set the update
frequency and press play.
If you have set up any
groups then new items
that match the search will
be automatically added to
the group and have any
colour or shape over-
rides applied.
TwitteractivityoverLondon
23. Bookmarks & Screenshots
At any point you can press
F12 to save a screen-shot to
your hard-drive as a PNG
image file.
You can also select Add
Bookmark to capture your
current location and view, and
then return to that point in the
future by selecting the
bookmark from the list on the
Camera tab. Multiple
bookmarks can be created,
and used to present your
analysis back to others by
walking them through it.
The Camera tab also lets you
control camera speed and
type.
VASTSimulatedSocialGraph
24. Examples
On the next few slides we'll show you some
examples of different visualisation types
created using Datascape.
You can see video examples, and details of
how the visualisations were created,on our
web site at www.daden.co.uk/datascape.
VASTSimulatedSocialGraph
25. Line Graphs
Datascape allows conventional line graph data plotting, eg for scientific and financial
data analysis, but in 3 dimensions. The Generate Network function can be used to
connect data points, and multiple panels can be brought in to segment the data, for
instance in to quadrants or octants.
UKGDPvsFTSEIndexvsBankruptciesover20years
26. Bar Charts
Datascape can create 3D bar charts, and you can decide whether you want to have
gaps between series (as here), or to have no gaps between them – giving you
something more like a surface graph. One advantage of the 3D chart over its 2D
equivalent is that you don't lose detail or relative scale if you have some bars which
produce significantly higher values than others.
InternationalCurrencyConversionRates–17currencies,20years
27. Bubble Plots
Bubble plots work really well in 3D, and in fact whenever you make size depend on
a variable you are creating a type of bubble plot. As shown here you can also use a
3D shape from the Datascape library as a reference for the plot, and future versions
will let you import your own 3D models.
Simulatedtraumaandcanceroccurrencedataover10yearsbybodylocation
28. Scatter Plots
Scatter plots are commonly used to represent scientific and other research data
where the relationship between variables is not yet know. By using three dimensions
and letting you use shape, colour and size in more powerful ways you can create far
higher dimensionality graphs than with most 2D tools, helping you to analyse data
more quickly, and to detect more complex relationships.
Systemsbiologyresearchdata–notyetpublished
29. GeoTemporal Plotting
A special case of the scatterplot is the geo-temporal plot, plotting Latitude and
Longitude on the X and Z axis, and time on the Y axis. Such data could vary from
scientific and business measurements, to live geo-coded social media feeds such
as Twitter.
TwitteractivityoverFarnborough,heightistime
30. Spherical Plots
Here we have plotted data onto a set of spherical axis, and then placed a globe in
the centre. This is also a geo-temporal plot since height above the sphere/globe is
again dependent on time.
Eurovisionsong-contestTwitteractivity,colouredbyArtist
31. Timelines
Another common format is the time-line, where one dimension represents time, and the other
axes are more abstract values (rather than geographic position). In this graph time moves
away from the user into the screen. Each dot representing one computer transaction, and
each line of dots a single computer having multiple transactions over time.
Computerfirewalldata,timeintoscreen,IPacrossscreen,colouredbydatadirection
32. Network/Social Graphs
Network graphs can range from the simple to the very complex. Layout can be specified in
the import file, or generated by the mappings. As well as social graphs you can also use
network graphs to map relationships as diverse as the links between proteins and the
topology of a computer network.
33. Other Graphs
These are just some of the most common graph forms that you can create in Datascape.
Since we give you complete freedom about how you map data to the plot features, and a
range of maths and string functions to use in those mappings, you are free to create whatever
graphs you can conceive and script. We will be publishing the mappings for our favourite plots
on our wiki and forum, and encourage users to do the same.
Computernetworklogdata,24hrspercircle,dayperspiral
34. Evidence Base
Prior to building Datascape we undertook an
MOD sponsored research project with Aston
University to compare immersive with non-
immersive 3D spaces for data visualisation.
The results (summarised right) were
encouraging, showing better performance in
an immersive space on the majority of
measures.
We have also published a White Paper on
Immersive Visual Analytics, which not only
provides more detail of this study, but also
provides a literature review of 2D and 3D
comparison studies.
Whilst 3D wont improve every visualisation,
the indications are that it will improve many,
and for the first time Datascape makes it
really easy to evaluate 3D for your business.
35. Advantages & Benefits
We use the Benefits Linking diagram above to show how the key features of Datascape link
to its main advantages over 2D and non-immersive 3D techniques, and how they in turn
can be used to drive out soft and hard benefits to your business.
36. Try It Now!
Datascape is available as a FREE fully-
functional 30-day trial download from:
http://www.daden.co.uk/conc/datascape
/datascape-download
Licences are available for corporate
users, and at a discount for educational,
SME and solo users.
There are videos and more information
on Datascape available on our web site
at www.daden.co.uk/datascape
37. About Daden
Immersive 3D learning and visualisation specialists
Software and hardware based immersion
Founded 2004, but 3D experience since late 1990s
Award-winning:
US Federal Virtual Worlds Challenge winner 2010
Finalist 2013 Global Unity3D Awards
Times Higher Education Winner 2009 & 2013
Research-inspired:
Member MOD Dstl Synthetic Environments Tower
Over a dozen MOD and InnovateUK R&D Projects
Based in Birmingham UK
We (and others) always felt that 3D immersion was better – but MOD funding allowed us to do a limited qualitative study. We hope to follow this up with a fuller study in 2012/13.
We (and others) always felt that 3D immersion was better – but MOD funding allowed us to do a limited qualitative study. We hope to follow this up with a fuller study in 2012/13.
We (and others) always felt that 3D immersion was better – but MOD funding allowed us to do a limited qualitative study. We hope to follow this up with a fuller study in 2012/13.