The Data Visualization Catalogue, helping you find the right data visualization method for your data. The Data Visualization Catalogue is a project developed by Severino Ribecca to create a library of different information visualization types.
Below is a guest post from Andy Maggio, the founder of DataView VR. He’s here to explain his vision of the future with dashboards and workspaces from utilising virtual reality (VR) technology. Andy will also demonstrate how VR can be useful in multi-dimensional data analysis and in the financial industry.
During my 14 years at Morgan Stanley as an investment analyst, by far my biggest challenge was trying to process the incredible amount of data and information that was available to me daily. Even though I had charts and dashboards at my disposal, it was still tough to make sense of it all. Especially when it came to multi-dimensional data.
Then in around 2016, I first started using VR hardware and from there I had two thoughts: first, that VR is going to be the most revolutionary technology of my lifetime; and second, that VR can make the process of data analysis and presentation much easier (especially in my job as an investment analyst).
Specifically, I believed VR could make data analysis faster, more effective, significantly improve my information organization capacity, and make the presentation of my analysis much clearer and memorable. Also, not to mention the fact that VR can drastically cut travel expenses by allowing you to work remotely.
So over the last two years, I put together a team that has endeavoured to create a product that accomplishes this vision.
The end result was DataView VR, a comprehensive VR based workspace, which harnesses VR, AR and AI in an immersive space that is designed specifically for analysts to provide them insights into complex data, improve productivity, and facilitate virtual collaboration. I believe DataView VR is the next step in the evolution of analysis and presentation software.
DataView’s “Workspace of the Future” is a virtual desk space that provides a combination of two-dimensional and multi-dimensional tools and displays in order to facilitate deeper, faster analysis.
I’ve implemented DataView in my own work and find it an excellent way to organize investment information, do data discovery and create predictive models. I feel it has both saved me time and allowed me to come to more profound conclusions. That being said, I did design the product to specifically address frustrations I had as an analyst.
Of course, when we speak with potential users, inevitably we are asked about specific applications of this platform. How can VR actually improve investment analysis?
Well, let me demonstrate to you a couple of examples of how DataView VR users are currently leveraging the platform.
In DataView VR, the “HyperDesk” is a saveable 18-monitor display that provides a great way to organize displays and then refer back to (and present) information about specific topics.
DataView VR Mult-Screen HyperDesk with Virtual Connectivity - YouTube
For example, an investor might have a saved HyperDesk configuration that is solely devoted to the Chinese Economy. It would include live feeds showing various relevant performance charts, custom charts, local news feeds and social media.
Over the course of the day the investor can also jump from the “Chinese Economy” room to a “Eurozone” room configuration, then after to a “Portfolio Performance” room showing price charts for all active positions. This allows for flexibility when working and is something impossible to do with a physical desk.
Give Presentations Anywhere
VR also opens an entirely new frontier of communication and presentation. For example, perhaps a client wants an update on an analyst’s work on a specific stock. The analyst can simply invite the client into the DataView VR Hyperdesk and have an immediate, face-to-face interaction, showing off research that combines both 2D and multi-dimensional analysis.
In the videos below are examples of specific applications of the platform when displaying multi-dimensional data. Using video is a pale representation of an immersive VR environment, but it can still give you a sense of the functionality. However, bear in mind that VR is really something that needs to be experienced to be fully understood.
Application #1: Searching for investment ideas based on multiple criteria
Here, each object in a visualization can represent a stock, and the characteristics of that stock communicate whether that stock is cheap vs history, whether its margins are depressed, whether it has strong earnings growth, whether it has a high dividend yield, what sector it is in, whether it has excessive leverage, etc.
It’s very intuitive for an investor to search for graph objects that are tall, fat, in a certain quadrant, of a certain color, etc. (in order to find the combination of criteria that they are looking for). DataView VR also allows users to quickly filter data ranges and drop outliers to expedite the data discovery process.
Application #2: Creating and visualizing multi-variable relationships, which is particularly useful in creating predictive models
To build a model to predict US corporate earnings growth, a user can input history for several dozen macro factors into DataView. After, a few things will happen:
DataView’s AI engine will automatically tell display 3-6 variables that have best predicted US earnings growth (after allowing the user to impose any restrictions on the data).
The user can actually see the output of the model in multiple-dimensions.
DataView provides relevant regression statistics related to the model and relevant 2D charts for reference.
The multi-dimensional visualization aspect of this process is powerful because there are certain investment relationships that are very important, but that simply can’t be seen unless it is possible to view more than two variables simultaneously.
DataView VR Multi-Dimensional Data Visualization - YouTube
VR drastically increases the effectiveness of multi-dimensional displays, which are very difficult to work with and understand on flat screens.
In this US corporate EPS model example, the USD dollar is an important historical driver of US EPS growth. However, in a 2D chart of the USD YoY vs EPS YoY like the in scatterplot below, you can’t see as much of a relationship.
In contrast, a VR multi-dimensional display that includes US GDP growth, the USD, and other macro factors, allows a user to very clearly see that a weak dollar drives stronger earnings when controlling for economic growth.
You can see that trees move from bottom left to top right because EPS growth correlates with GDP growth (EPS on X, GDP on Y). To the right of this trendline you can see a string of skinny trees: the skinnier the tree, the weaker the dollar was in that quarter. The weak dollar shifts the EPS/GDP relationship to the right, showing that a weak dollar boasts EPS growth.
This type of visualization allows decision makers to more confidently act on their predictive models, as the model becomes something that can be seen rather than just an equation on a page.
In a multi-dimensional display, a user can quickly see many aspects of the performance of a portfolio of stocks.
DataView VR Multi-Dimensional Data Visualization - YouTube
For example, for each stock in the S&P 500, performance can be overlaid simultaneously with several factors, providing a quick answer to questions like:
What’s been driving out-performance/under-performance?
Is EPS growth relevant right now?
Is a market cap a driver?
Is performance concentrated in specific sectors?
Is momentum important?
Are investors hiding in safe stocks?
And so on.
I strongly believe that VR, AR and AI are going to revolutionize the way that people interact with data and information over the coming years. So right now I’m working hard to make this transformation a reality for investment professionals like myself.
This is the third post looking into 3D Treemaps. This time we’ll be exploring 3D Treemaps that use extrusion to communicate an additional variable.
If you’re familiar with CAD or any 3D drawing software, you’ll be aware of the ability to make flat shapes into 3D by ‘extruding’ them with a certain tool or menu option.
3D Treemaps that use simple extrusion with rectangles are maybe easier to understand in comparison to the last two types of 3D Treemap design (Cubes, Spheres and Cylinders). This is probably due to it being easier to understand and interpret that the longer the length of extrusion, the high the value.
While many of these 3D Treemaps covered here or in the previous posts may not be considered good design because of the issues with displaying charts in 3D. Despite this, I don’t think we should ignore them.
So instead, there will be a showcasing of what’s been done and what is possible. Think of this more as a display of xenographics – charts that are weird and unusual, but are also interesting because of this.
However, as you will see in this post, some of the 3D Treemap examples have some practical applications to them.
Source: Visualization of Software and Systems as Support Mechanism for Integrated Software Project Control, p. 6, Fig. 3
This chart bears resemblance to a 3D Bar Chart, but of course, functions differently. No detail on the chart’s design is given in this study, but it does demonstrate using extrusion to display an additional data dimension on a Treemap.
With this 3D Treemap, there doesn’t seem to be any nesting of subcategories in the design there. Creating to a more jumbled and chaotic display of blocks.
However, one 3D Treemap design that does nest its subcategories is StepTree, which does this by stacking each subcategory on top of the parent category. StepTree was developed in the paper Extending Tree-Maps to Three Dimensions, A Comparative Study. The intention behind this study was to investigate ways of enriching space-filling visualisations so that size would be less dominant.
From their research, the authors of this study found that StepTree had longer task times in comparison to a 2D Treemap. However, in the display of hierarchical depth, the StepTree displayed a clear advantage.
Here, a visualisation was created by readapting a Treemap’s layout to be more suitable for showing the structure of software. The reason behind developing such a visualisation was for analysing the software’s coding quality and for aid in the understanding of large-scale software systems.
Source: Visualization-based Analysis of Quality for Large-scale Software Systems, p. 5, Fig. 5 (top)
In this adove visualisation, 3D boxes are used to represent a ‘class’ in the programming. The box height represents the code size. Colour is used to display coupling, so red represents high-coupling, which is bad in an object-oriented paradigm. Rotation represents the lack of cohesion, because the twisted boxes look more chaotic.
The end result produces a sort of ‘city’ metaphor, where the classes can be viewed as building and the software packages as districts.
Using the city metaphor for visualising software is explored further in the paper Visualizing Software Systems as Cities, which describes these charts as a ‘CodeCity’. The goal here is to ease comprehension of software systems by representing them as a city that can be traversed and interacted with. The idea is that the sense of locality would aid in making the software system easier to understand.
Source: Visualizing Software Systems as Cities, p.3, Fig. 2
Here, 3D Treemaps were tested for visualising the classificatory distribution of patent collections in the International Patent Classification (IPC) system. The focus of this study was to use advanced interactive visualisation techniques to aid in the analysis of massive patent collections and portfolios.
Source: A 3D Treemap Approach for Analyzing the Classificatory Distribution in Patent Portfolios, p. 2, Fig. 3
The 3D Treemaps used in this study used the third dimension to represent the number of patents associated with a particular category.
In addition to that, there is also an example of a “Stacked 3D Treemap”, which stacks multiple patent sets per category upon each other and colour-codes them (using red and blue in the example below).
Source: A 3D Treemap Approach for Analyzing the Classificatory Distribution in Patent Portfolios, p. 2, Fig. 4
From exploring all of these examples, you can see that this visualisation isn’t just a wacky experimental way of bringing a Treemap into the third dimension. 3D Treemaps that use extrusion are popular in the visualisation of software systems. Not only has this been discussed in academic papers, but there are a couple of commercial products out there that have implemented this kind of visualisation. Also, others have used 3D Treemaps in other applications such as in the analysing of vast patent libraries. With virtual reality and augmented reality becoming more popular, maybe 3D Treemaps could be applied there and possibly function better than on a flat screen.
Interactive Poster: Exploration of the 3D Treemap Design Space, H. Schulz, M. Luboschnik, H. Schumann
Recently, I launched an eBook guide on using the built-in Graph Tools in Adobe Illustrator.
From this eBook you can learn to master the Graph Tools for infographic and data visualization work through step-by-step tutorials and working exercises.
The eBook also teaches you all the advanced techniques not normally thought capable on Adobe Illustrator, like combining two chart types together or adding code to bits of text to program in value labels.
The eBook also contains tricks to exploit the Graph Tools to product additional charts, such as Slopegraphs and Bar Code Plots, which are not normally available with the built-in Graph Tools.
How to Create Slopegraphs in Adobe Illustrator - YouTube
There’s also a section on how to make charts responsive for the web, by using the ai2html plugin.
However, I can understand that live online training isn’t for everyone. So instead I’ve provided a section of the courses can people can go through in their own time at a fraction of the cost.
Why would I use Illustrator to draw charts when there are better tools out there?
While it’s true that there are many more powerful and advanced tools out there for creating pieces of data visualization, they’re not for everybody. Some require coding skills, while others require training.
The guide I created is more for those who already know how to use Adobe Illustrator, but they don’t know how to fully utilise the Graph Tools for their, typically, infographic work. I wouldn’t suggest Illustrator to those looking to create dashboards. But Illustrator is a tool best suited for creating infographics.
You can use other tools like Excel and copy + paste them in. However, sometimes the charts are not completely transferred over or corrupted and there’s some tweaking involved. If you know how to use the Graph Tools properly, it’s faster to use them to create the charts inside Illustrator instead.
Adobe Illustrator is also great for creating custom charts, as it gives you full control over the visual aspects of a chart’s design.
Where can I get a copy?
You can purchase a copy on the store here. The eBook contents are also displayed there.
The eBook comes in both PDF and EPUB format, is 37 pages long, and fully illustrated with instructional screenshots.
So in the previous post in the series, I covered Treemap visualisations that use a 3D cube shape to displays hierarchal data. However, there are still a number of different 3D Treemap visualisations that I need to cover, in particular, 3D Treemaps that used spheres or cylinders.
Source: The Design Space of Implicit Hierarchy Visualization: A Survey, p. 12, Fig. 6 (b)
A spherical variation was also tested out:
Source: The Design Space of Implicit Hierarchy Visualization: A Survey, p. 12, Fig. 6 (c)
3D Polar Steptree
Another experimental variation of a Treemap was carried out by Schulz and co, by taking the 3D Polar Treemap a step further by combining it with elements of the Steptree.
Source: The Design Space of Implicit Hierarchy Visualization: A Survey, p. 14, Fig 10
One of main limitations of a Treemap is that it more difficult to discern the hierarchal structure than a Tree Diagram for example. So to solve this problem, Junghong Choi, Oh-hyun Kwon and Kyungwon Lee developed their own variation of a Treemap, which they named the Strata Treemap.
[…]Strata treemap, a new approach to treemap presentation that im- proves the visibility of the hierarchical structure without distorting the node sizes. Strata treemaps are based on Voronoi treemap. Strata treemap use the surface of a sphere instead of the traditional two-dimensional plane. For each node in the hierarchy, we extrude it along the its surface normal and stack up on the its parent. The result is a three-dimensional shape that consists of stacked blocks on a sphere. Each block represents a node in a hierarchy and its top surface area represents the quantitative value of the node.[…]
~ Strata Treemaps, p. 1
Interaction techniques for a Strata Treemap - YouTube
[…]An attempt was made to “mash” visual techniques together: SphereTree attempted to combine a treemap e.g. [Joh91, Bru99] projected onto a sphere together with an internal hierarchy through the center of the sphere. The hierarchy was presented as a successive series of concentric shells, with each treemap not completely filling its area, leaving gaps to view each successive underlying shell. Difficulties with visually associating patches of the treemap with the corresponding parent/children within the hierarchy led to some iterative exploration and adjustments, eventually settling on replacing the inner shells with a ball-and-stick hierarchy inside the sphere.[…]
~ Sphere-based Information Visualization: Challenges and Benefits, p. 4
Source: Sphere-based Information Visualization: Challenges and Benefits, p. 4, Fig. 11
Continuing on at looking into the claims being made about virtual reality (VR) data visualization, we will now look at the second most commonly made claim:
VR data visualization is great for multidimensional data analysis; VR allows you to view the data in ‘multiple dimensions’
From taking a guess, we can assume that this relates to VR’s ability to display data in 3D. So the additional dimension here would be depth (z-axis). But because we don’t perceive depth well, due the distortions and other issues that can occur, would having an extra dimension be of much benefit?
If VR is capable of displaying more than one extra dimension, how would this be beneficial? In Stephen Few’s critical post on VR data visualization, he makes the point that:
[…]And let us not forget that we can only hold from three to four chunks of visual information in working memory at once, so even if VR could add many more dimensions of data in some way, it would be of no use to our limited brains if we weren’t able to process all of those dimensions simultaneously.
However, from my research online, this is what I’ve heard being said about the VR’s ability to visualize multidimensional data:
Okay, let us talk about living in the real world again. There are five senses that determine our existence in the real world, but in a typical scenario, we could only use our sight to analyse and interpret data. But what if we could use other senses too? This is exactly what Virtual Reality offers in data visualization. For example, take the case of hearing. With data-audio relationships, we could easily determine the location, subject, and significance of a specific data through its direction, loudness, and type! With haptic feedback gloves gaining popularity, we are not far away from a period where we could actually feel the data.
By using multiple senses, we can enhance our ability to process data with more dimensions. While it may be a bit radical to talk about taste and smell in a data visualisation context, it is not outside the bounds of possibility to ‘feel’ data. This is technically achievable right now with haptic feedback gloves.
Before VR made its entry, users had only their eyes to analyze the data, but post VR, multi-dimensional data analysis is possible. And that means, not just the use of hands, but hearing as well. This enables them to understand the subject, location and significance of a particular data source.
Maybe it is a little out of bounds to talk about feel and taste in data visualization, but perhaps, such a day is not too far behind, right?
So from all this we can see that the claim of VR allowing you to view the data in ‘multiple dimensions’ doesn’t just relate to depth or another form of visual communication. VR also allows you to process data in the form of sound, touch, smell and soon in the near future, even taste.
This Mask Allows You To Smell Virtual Reality - YouTube
Maybe we’ll be able to tell if the data smells bad?
Still, I don’t see how this would aid in data analysis. But in terms of communicating the data and a narrative, I could see this being really novel and engaging. Not everyone learns well visually and are instead better at absorbing information through different means, for example, by auditory or kinesthetic learning.
To get some industry insight into the topic, I spoke with Suzanne Borders from BadVR. Here’s what she had to say:
Visualizing univariate data in 2D is pretty straightforward but issues begin arising when the number of dimensions in the data increases. These issues occur because most data visualizations are bound by the 2D visual displays used to present them.
Additionally and unfortunately, as modern datasets trend towards ever greater scale and complexity, it becomes more and more difficult over time to visualize multivariate datasets in a way the accurately communicates all the existing depth and complexity.
Human analysis on 2D screens require multiple different summary charts, graphs, and reports which must be manually cobbled together by humans to inform decision-making. Obviously, this method has it’s downsides. Complex interrelationships are impossible to find, and due to sheer scale, a large amount of the dataset remains unexamined from certain angles.
If you remove the compromises required by 2D screens, a whole slew of new ways to visualize data become possible. Users can view one big picture of their entire dataset and literally ‘step inside their data’ to see and analyze in real-time. VR also offers the ability to massively increase the scale of data presented to the user, allowing users to see everything all at once, removing the need for summary reports or charts.
Abstracting large datasets down isn’t necessary when you have depth, position in space, position to other data points, texture, animation, and many more ways to communicate multiple data attributes, and a limitless environment in which to do so.
There’s also really cool video from Project NEO by Francois Bertrand, which demonstrates VR’s usefulness in viewing multidimensional data to detect patterns. There’s even a feature to view the data in a 6-dimensional hypercube:
Project NEO: Virtual Reality Data Visualization for Machine Learning - YouTube
Although this study doesn’t include VR technology, it does show how that even using 3D visualization on a screen yields better results on particular visual tasks. For example, when it came to distance perception, the error values produced in the 3D version are much lower than those in 2D (p. 13 – 14). Overall, it was concluded that there was a significant loss in quality when switching from a 3D to 2D visualization.
. . .
In summary, it’s pretty obvious that VR allows you to visualise multidimensional data, not only by using depth and animation, but also in non-visual forms of communication such as sound, touch and smell. While all of these features are not unique to VR, I would also argue that the combination of all the other features into an immersive experience is the unique trait of VR. While there seems to be some evidence to suggest that VR is useful in the display and analysis of multidimensional data, more research needs to be done.
I’d also like to give a special thanks to Andrea Bravo for the academic suggestions and insight.
New insights into the suitability of the third dimension for visualizing multivariate/multidimensional data: A study based on loss of quality quantification, A. Gracia, S. González, V. Robles, E. Menasalvas and T. von Landesberger
DataViz: visualization of high-dimensional data in virtual reality, E. Feng and X. Ge
Virtual Reality-based Human-Data Interaction, E. A. Widjojo, W. Chinthammit and U. Engelke
Originally, I was planning to write a Further Exploration post on just Treemap variations. I thought there would only be a small, handle-full of charts that I would need to cover.
But man, I was wrong.
Once I actually sat down and begun researching into other Treemap types out there, I found myself surprised and overwhelmed.
This was largely thanks to the work done by Hans-Jörg Schulz in his treeviz.net project. If you haven’t seen this site before, I’d recommend you checking it out. It’s a library of charts that specialise in visualising hierarchal or tree data.
Anyway, instead of doing a single post on Treemap variations, which would be way too long, I’ve instead opted to break down the charts I’ve found into more digestible posts. Starting with of course, on 3D Treemaps.
This type of 3D Treemap uses a cube design that has multiple layers within it, like an onion.
Since we live in a 3-D world our point of view is necessarily “within” the partitioning space of higher dimension treemaps (3 or more partitioning dimensions). Higher dimension treemaps require transparent or translucent nodes in order to make internal partitions visible from external points of view. Point of view can be interactively controlled and users can fly through 3-D partitionings but global perspectives require external views of the entire hierarchy.
Treemaps: Visualizing Hierarchical and Categorical Data, Johnson, p. 42 – 45
Here, transparent cubes are nested inside one another, and category labels are also displayed in 3D. Much of this design was done with virtual reality technology in mind, as various 3D interaction techniques were provided and taken into consideration.
Source: The Information Cube: Using Transparency in 3D Information Visualization, p. 127, Figure 2
A decade later, Y. Tanaka, Y. Okada and K. Niijima would also explore how to bring a Treemap into the 3D realm by using a cube design. In their paper Treecube: visualization tool for browsing 3D multimedia data, they attempt to show the usefulness of visualising a Treemap for browsing 3D multimedia such as 3D shape models and motion data, but also for 2D image galleries as well.
Source: Treecube: visualization tool for browsing 3D multimedia data
Another 3D Treemap cube I came across, but couldn’t find too much information on was the “Magic Treemap Cube” created by Xiaowu Chen, Haolin Yang, Yongtao Ma and Bin Zhou.
[…] we present a novel extension called magic treemap cube, which improves the layout and has a good performance on visualizing unfixed (cross-level) and complex data, inverting hierarchy without reorganizing the structure, and enabling users to easily compare among treemaps. […]
International Journal of Virtual Reality; 2010, Vol. 9, Issue 3, p. 9
As you can see below, this design somewhat resembles a Rubik’s Cube. The Magic Treemap Cube was apparently used in an application called OlympicViz to visualise data for the Beijing 2008 Olympic games.
Initially, I thought to myself, how could virtual reality actually make things seem more natural? The fact that it’s called VIRTUAL reality already implies that it’s something unreal and not connected to the natural, physical world. Being in a VR world is completely unnatural because you are immersing yourself in a digitally-constructed reality.
But after going through my research, here’s what I found being said on the topic:
What helps us connect with our environment in the real world? It is the ability to easily touch and feel the objects around us, isn’t it? But what happens in a digital world? For long, we have relied on mice and keyboards to facilitate this sense of interactivity. But now, with virtual reality, we can gain a more natural sense of interaction by physically pushing buttons, moving the objects around and by even controlling data streams. So, just as you walk around your sofa or computer table, you can literally move around huge volumes of data. Pretty cool, uh?
[…] The immersive nature of VR, combined with the more natural presentation of information in VR, allows the user to increase the amount of data they process. […]
People can use VR to step inside the massive data sets generated by the business to observe them in a more intuitive and natural way. In doing so, it allows them to process this information more efficiently without going into cognitive overload.
The possibilities offered by VR are huge, and nowhere is it more important to businesses than data visualization. The naturalistic way it allows people to interact with virtual objects, the ability it affords individuals to experience the data through all senses – touch, sound, and even smell – mean that there are many opportunities yet to be explored, and data visualization companies should start preparing.
We use our hands to feel things around us; virtual reality allows doing this through data visualization. Companies can manipulate data streams, push windows around, press buttons and practically walk around data worlds with the help of VR technology. Consequently, it enables users to make accurate data analysis and facilitates faster decision-making process.
Virtual Reality allows us to touch an information as a real things with the help of data visualization. Today it is possible to manipulate data directions actually walk around data worlds. This allows users to make accurate data analysis and thus, faster decisions.
In the real world we interact with objects directly with our hands. This allows us to connect with the environment around us and get a better idea of the objects we’re dealing with. For a long time, we’ve used keyboards and mice as conduits for this interaction. Through virtual reality, we can return to a more natural way of interacting — by physically pushing buttons, moving windows around and manipulating data streams (such as in this VR assisted biological specimen analysis). That is in addition to being able to walk around and through these data worlds.
In the world that we live in, we can use our hands to feel things. Virtual reality allows us to do the same thing through data visualization. You can manipulate data streams, push windows around, press buttons and actually walk around data worlds. This allows users to make accurate data analysis and thus, faster decisions.
I was introduced online to a PhD researcher, Andrea Bravo, who I talked with to better explain to what is meant by having more natural interactions in VR. Andrea is currently researching into data visualization in VR based on the experience of “being immersed in data” and how does this affect decision-making.
From our conversation, Andrea explained that VR enables high-end user computer interfaces that replicate a scenario with high fidelity, through multimodal input. Also that having different input modalities and also different sensory feedback allows for more natural interaction.
Andrea went on to clarify that by ‘natural’ she understood in regards to the incorporation of our other senses in the experience because that’s our natural way of perception (as we have five senses). VR allows for exploration of data with the incorporation of those other senses, such as sound but also haptics (touch).
Andrea also went on to explain the importance of using haptics and tangible objects for natural interaction. As the fact of being able to use the two hands for data exploration also allows for natural interaction.
In the paper The Hologram in My Hand, there is some evidence of this with visualizations shown on some tests comparing desktop data visualisations with tablet AR and immersive tangible AR with the Hololens. With the HoloLens, users are about to “touch” the data by reaching with their hand inside a hologram visualization.
To elaborate further on what is meant by natural interaction, Andrea defines natural interactions as behavioural components close to the way we interact in real-world environments.
She goes on to explain that in the real world we have certain manners of interaction. But when in-front of a desktop computer this interaction transforms completely because we use a keyboard and mouse to interact with the world on the screen. If you want to move something on screen you just press the mouse button or use the keyboard, but this is not how we interact with objects in real life.
However, in VR, you can interact with objects by grabbing them and moving them around with your hands, as you can do in physical space. There is even the option for special gloves that give haptic feedback when interacting with objects in VR.
So everything about VR allowing for more natural interactions is true. However, this doesn’t necessarily mean that it will aid in the visual communication of data. In the future, it would be good to see more research on how having more natural interactions with data visualization aids in the analysis and reading of the data.
The Hologram in My Hand: How Effective is Interactive Exploration of 3D Visualizations in Immersive Tangible Augmented Reality? B. Bach, R. Sicat, J. Beyer, M. Cordeil & H. Pfister
Virtual reality-based human-data interaction, E. A. Widjojo, W. Chinthammit & U. Engelke
An investigation into the implementation of virtual reality technologies in support of conceptual design, R. I. Campbell, T. Page & K. S. Badni
An advanced virtual reality multimodal interface for design, R. De Amicis, G. Conti & G. Ucelli
That’s right, the chart icons from the homepage are now featured onto a mug. So now you view through a list of chart types, while drinking your favourite hot beverage.
This sturdy mug is made from quality white and glossy ceramic. It’s perfect for any data or tech enthusiasts out there who like to drink their coffee or tea while working with data or for simply relaxing. Here’s my own cup of ginger tea:
The capacity of the Data Visualisation Catalogue Mug is 325ml (11oz ) and it is completely dishwasher and microwave safe, so you don’t have to worry about damaging the design.
So if this is the mug for you or you think it will be a great gift for your data geek friends, then head over to the store to make an order. We stock in both the USA and Europe, but the mug can be shipped Worldwide.
I’ve also recently designed the front cover for a notepad that you can also get here.
You may or may not have noticed some changes to the store. It’s been neglected for some time now, but recently I’ve done some updated to improve it.
I’ve made some tweaks to interface/design. Mainly there has been some adjustments in the header, both in the title and menu, and also in the sidebar.
To make the store items more accessible to rest of the World, I’ve added 6 new currencies:
Two of the posters I designed years ago: the Greek Myth Family Spiral and Norse Mythology Family Tree have also been added in both poster and framed poster and in a few different size formats for each:
For two years in a row now, I’ve published what chart reference pages have been the most popular on the website.
In the initial post in April 2017, I looked at this data from the launch of the website (26th December 2013) till 1st February 2017 (so around 3 years worth of data). While, the post from last year looked at the data from 6th Jan 2017 to 6th Jan 2018, which only showed the page views from a single year.
So like in the previous year, I will only be looking at the page views data from the past year (2018).
Also, even though I’ve launched a few different language versions of the website, I will be only using the analytics data from the English website. The number of page views shown is for the entire year (start of Jan 2018 to Jan 2019).
Now without further ado, here are the top 10 chart reference pages for 2018:
Below I’ve displayed in a table the change in rankings and page views between 2017 and 2018. Charts that have increased their rank are highlighted in green while those that have dropped in rank are coloured in red.
This year there have been a few dramatic shifts in ranking. The Box Plot has completely dethroned the Choropleth Map, which doesn’t even come in at 10th place (it was knocked down to 11th place)!
The Radar Chart was also knocked out from the top 10. While the Histogram managed to catapult itself onto 9th place all the way from 19th place with a whopping increase in page views of 105.9%!
But across the board, all the listed chart reference pages have all increased their number of page views over the year.
I drew a Slopegraph visualising the difference between page views between 2017 and 2018, as this chart better displays the gap between values and provides more detail.
Here we can clearly see a dramatic increase over the year for the Box Plot (154.9%), Density Plot (76.3%), and Histogram (105.9%). The Box Plot reference page has not just reached first place, but it skyrocketed there with a massive 44,807 page views that year.
All three of these charts that have seen a large increase in their viewership are used to visualise data distribution. Maybe that’s some clue to why they’re popular this year?
The rest of the chart reference pages are clustered close together in both years and have only had a yearly increased ranging between 24.3% to 39.6%.
Anyway, it’s been interesting looking through this data. Maybe next year when I have data for 3 years (2017, 2018 and 2019) I can then use a Bump Chart to visualise the shift in ranking over those years.
When combining an Arc Diagram with another form of data visualization, the additional graph markers can be placed on or aligned with the connecting nodes of an Arc Diagram.
This allows for an additional variable (or multiple variables) to be added to the diagram. This additional variable is of course connected (or relevant) with the node data.
So for example, if an individual node represents a single person in a network, the arcs represent who they’re messaging, and then the additional variable marker (say a bar) attached to the node has a length proportional to the number of messages that user has sent.
+ Bar Chart
First up is an illustration of the previously presented example: a combination of a Bar Chart with an Arc Diagram.
Working together with pastor Christoph Römhild, Christ Harrison produced an Arc Diagram that visualises biblical cross-references. Here, each arc represents the 63,779 cross-references found in the Bible. Chris also combined this diagram with bars at the bottom, which each represent a chapter in the Bible with the lengths corresponding to the number of verses contained within it.
Source: Bible Cross-References, Chris Harrison
+ Stacked Bar Chart
Of course, if an Arc Diagram can be combined with a Bar Chart, you can be sure that a Stacked Bar Graph can be applied as well.
The 100% Stacked Bar Chart variation can also be applied. In the example below, the stacked bars have been fixed to have a uniform total width as a way of displaying part-to-a-whole relationship at each node.
Another way to display discreet, numerical comparisons could be through the use of a Unit Chart. Here, a shape is used to represent a numerical unit (for example, 1 block = 10).
+ Pictogram Chart
Instead of using simple shapes, you could instead substitute them for more detailed icons and uses a Pictogram Chart to combine with an Arc Diagram.
+ Span Chart
Another way to use bar markets on an Arc Diagram could be through the use of a Span Chart – especially if you’re looking to display ranged data. Here, the bars are not attached to each node, but are aligned to them. An axis could also be displayed along with the chart to give readers the ability to read values.
And if you want to have two variables or display sections of the ranges or show a part-to-a-whole relationship of the ranges, then the bars here can be split.
+ Proportional Area Chart
Numerical quantities can also be displayed fairly well through the use of shape areas (like on a Proportional Area Chart) on each node.
The example below from Martin Dittus visualizes data from IRC (Internet Relay Chat) communication behaviour to see who is speaking to whom, and who is namedropping whom. Here, the size of the circles represents the number of messages sent by each user. The varying color shades of the circles is based on the average message length.
Source: IRC Arcs
If unlike in the example above, your Arc Diagram doesn’t have arcs on the bottom the top and the bottom, then an alternative design could be to use semi-circles on the bottom half to reduce clutter and overlapping of the shapes areas on the arcs.
+ Pie or Donut Chart
Replacing nodes with Pie Charts or Donut Charts is a straightforward way to display percentages, although they’re not so fantastic to compare across.
The example below is part of a chart by Gaston Sanchez, which visualizes the script from the original Star Wars trilogy. Here, the semi-circle Donut Charts have been used to show the amount each character speaks across the three movies. The size of each semi-circle Donut Chart is also proportional to the amount of dialogue that character has in total.
Source: Star Wars Arc Diagram, Gaston Sanchez
If having arcs connecting to other arcs is too much for you, then an alternative design for this chart combination would be to flip the ‘arches’ onto the bottom.
+ Waffle Chart
A clearer way of displaying and comparing percentages could be to use a Waffle Chart under each node.
To combine an Arc Diagram with a Heatmap, all you need is to shade shapes near or on each node. Multiple data series can be displayed with additional rows of shaded blocks.
+ Line Graph
It’s possible to combine an Arc Diagram with a Line Graph, by aligning each point to a node and varying the vertical positioning based on a value axis.
Despite this possibility, this chart combination isn’t actually visualizing data over time (unless each node represents a series of uniformly distributed points in time).
+ Area Graph
Combining with an Area Graph is also another possibility. But like with the previously mentioned Line Graph combination, this isn’t necessarily showing data changing over time and is most likely just showing the data values increasing or decreasing over each node.
+ Word Cloud
If your network is connected with text-based data, then combining the Arc Diagram with varying sized text like in a Word Cloud is an option.
Here, you could display a few different variables on the text through the choice of word listing, ordering, and making the size of the words proportional to the frequency of occurrence or some other numerical variable.
This chart combination is useful for comparing bodies of text that occur within an interconnected network. For example, in a messaging network, where the arcs signify who is messaging who and the text visualization underneath the nodes shows the most commonly used words or phrases being exchanged.
A simpler way of displaying data distribution can be done by combining with a Dot Distribution Plot.
+ Barcode Plot
Another way to display data distribution with an Arc Diagram is to combine it with a Barcode Plot.
The example below by Matthias Dittrich contains an example of this combination. Here, all emails of a particular weekday were shown on a 24-hour time strip. The replied e-mails are represented by the arcs linking to the original e-mail source. The goal of this visualization was to show when and how fast emails were being answered.