Archive for May, 2011

The Future of Computing (Now and Then) – Visualisation

Visualisation, as the name implies, is about methods or techniques of displaying (visualising) data, in such a manner that the result is further and better understanding of the underlying information or knowledge of the data, what the data is “imparting” to the audience.

Visualisation is closely aligned with Analytics in many ways, in that Visualisation is all about how to present large volume and complex data in such a manner that it is (relatively) simple for humans to understand the underlying message, or pattern, represented or conveyed by the data. Typically, the analytics process will apply algorithms to process the data in a variety of manners, hopefully obtaining a suitable result – which then needs to be presented such that decisions can be made and further action can be taken. In many instances, the analytic process itself may involve Visualisation in order to allow the people performing the analytics to determine what to do next.

These insights into the data – what it is trying to tell us – are imparted through the power of the human mind, through its ability to make connections and understandings according to visual cues (and prior knowledge). Interestingly, the ability for humans to understand based on visualisation is mainly achieved through additional computing processing – advanced visualisation typically involves advanced information processing (and sometimes advanced hardware for particular visualisation purposes, such as 3D displays, large screen projections, etc). It should be noted though, that visualisation has been around for quite some time – early maps are a form of visualisation – producing a visual representation of some data to assist in understanding (see http://www.math.yorku.ca/SCS/Gallery/milestone/milestone.pdf).

In the past, Visualiation has been part of Business Intelligence (think charts, graphs, dashboards), but in a rather simple and simplistic manner. Modern visualisation uses all the power of graphics and animation (including 3D) to present a compelling vision for decision making. Visualisation today is much much more than a few charts and graphs. It is many different types of graphical representations; it is animated timelines (combined with multiple graphic types); it is now 3D (both static and animated) and it is interactive (the initial visualisation can allow those viewing to select an element, which will then query for new data (for instance, more detailed information which is then visualised (possibly using a different mode of visualisation) and allows for further interaction).

Some disciplines and specialist areas only operate based on visualisation. Areas include:

Computed Axial Tomography (CAT) and Magnetic Resonance Imaging (MRI) scans in medicine only “work” because the massive amounts of data generated are presented in a visual form for the specialist physician to interpret. The physician never works with the raw data – only with a computed representation (http://www.humansfuture.org/visualize_scientific_visualization.htm). Astronomers now regularly use visualisation to process the huge amounts of data generated by modern telescopes (current estimates suggest that this data stream will exceed 1 Terabyte of data per day in the near future – see http://astronomy.swin.edu.au/scivis/ and http://astrocompute.wordpress.com/2011/03/04/scientific-visualization-in-astronomy-towards-the-petascale-astronomy-era/) to visualise and understand how the universe works – and what it looks like, from a variety of perspectives – not simply in the human visual spectrum, but also in the infrared and ultraviolet wavelengths, as well as gamma radiation and other signals. Geospatial data, such as topography, hydrography, etc, are all presented in terms of visualisation – as simple as a mapping display, up to as complicated as real-time 3D animation through a timeline.

Visualisation today is enabled by hardware advances, specifically the inclusion of Graphical Processing Units (GPUs) in computers (specifically PC based hardware) – offloading the processing for visualisation from the standard CPU onto a dedicated and high-powered chip (see http://ieeexplore.ieee.org/Xplore/login.jsp?url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel5%2F9449%2F29999%2F01372245.pdf%3Farnumber%3D1372245&authDecision=-203; http://astrocompute.wordpress.com/2011/04/15/gpus-vs-cpus-apples-vs-oranges/ and http://astronomy.swin.edu.au/scivis/).

An excellent visualisation graphic of the different types of visualisations which can be produced, categorised into six different types (data; information; concept; strategy; metaphor and compound) has been produced by Visual Literacy (which provides e-Learning tutorials on visualisation – see http://www.visual-literacy.org/) and is available at http://www.visual-literacy.org/periodic_table/periodic_table.html. The web page is interactive, displaying an example of each type of visualisation when mousing over the entry box in the table. An excellent example of how to do visualisation well!

Some of the specific techniques used for visualisation include:

  1. a Cladogram (for display of phylogeny – see http://darwiniana.org/trees.htm; http://www.scribd.com/doc/51895193/6/Cladograms-and-Trees; http://cnx.org/content/m11052/latest/; http://www.crescentbloom.com/ii/l/17.htm; http://www.dinosauria.com/jdp/misc/cladogram.html);
  2. a Dendrogram (for display of classifications – see http://botanydictionary.org/dendrogram.html; http://www.nonlinear.com/support/progenesis/samespots/faq/dendrogram.aspx; http://www.mathworks.com/help/toolbox/stats/dendrogram.html);
  3. Graph drawing (http://graphdrawing.org/; http://gdea.informatik.uni-koeln.de/; http://www.ogdf.net/doku.php);
  4. Heat-maps (http://www.cs.uic.edu/~wilkinson/Publications/heatmap.pdf; http://www.patrick-wied.at/static/heatmapjs/; http://ashleylab.stanford.edu/tools_scripts.html; http://www.bioinformatics.ubc.ca/matrix2png/);
  5. Hyper Trees (http://www.sigchi.org/chi95/Electronic/documnts/papers/jl_bdy.htm; http://hypergraph.sourceforge.net/; http://thejit.org/; http://ucjeps.berkeley.edu/map2.html; http://www.touchgraph.com/navigator);
  6. Treemapping (http://www.magnaview.nl/documents/Visualizing_Business_Data_with_Generalized_Treemaps.pdf; http://www.perceptualedge.com/articles/b-eye/treemaps.pdf; http://www.cs.umd.edu/hcil/treemap-history/index.shtml; http://iv.slis.indiana.edu/sw/treemap.html).

In the future, such visualisation will link to differing models of Human Computer Interaction (HCI), including various forms of haptics (“The science of applying tactile sensation to human interaction with computers” – source: http://foldoc.org/haptics. Also see: http://www.immersion.com/docs/Value-of-Haptics_Jun10-v2.pdf) and immersive technologies, such as the multi-touch desktop (see http://www.perceptivepixel.com/ – similar to the technology hypothesized in the movie The Minority Report, based on the short story of the same name by Philip K. Dick).

No Comments

6 Easy Steps to Falling Asleep Fast

Another post from the PsyBlog, once again, re-posted with kind permission.
Something that we are all going to have to pay attention to at some time in our lives.

6 Easy Steps to Falling Asleep Fast

Post image for 6 Easy Steps to Falling Asleep Fast

Psychological research over three decades demonstrates the power of Stimulus Control Therapy.

Can’t get a good night’s sleep? You’re not alone. In surveys of what would improve people’s lives, a good night’s sleep frequently comes near the top of the list.

Poor sleep results in worse cognitive performance, including degraded memory, attention, performance and alertness. And in the long term insomnia is also associated with anxiety and depression. And people’s sleep gets worse as they get older. After 65 years old, between 12% and 40% of people have insomnia.

All sorts of methods have been tried to combat poor sleep, from drugs through psychological remedies to more outlandish treatments.

The problem with drugs is that they have side-effects and are often addictive. The problem with the more outlandish treatments is that although they tend not to have side-effects, we don’t know if they have any effect at all. Psychological remedies, though, combine the best of both worlds: studies show they work without side-effects.

Stimulus Control Therapy

Professor Richard R. Bootzin has been researching sleep disorders for many years at the University of Arizona Sleep Research Lab. Writing in the Annual Review of Clinical Psychology, he describes the different psychological approaches that have been used to treat insomnia (Bootzin & Epstein, 2011).

Of these the most successful single intervention is called Stimulus Control Therapy (Morin et al., 2006). You’ll be happy to hear it consists of six very straightforward steps. If you follow these it should improve your sleep. After the list I’ll explain the thinking behind them. First, here are their six steps:

  1. Lie down to go to sleep only when you are sleepy.
  2. Do not use your bed for anything except sleep; that is, do not read, watch television, eat, or worry in bed. Sexual activity is the only exception to this rule. On such occasions, the instructions are to be followed afterwards, when you intend to go to sleep.
  3. If you find yourself unable to fall asleep, get up and go into another room. Stay up as long as you wish and then return to the bedroom to sleep. Although we do not want you to watch the clock, we want you to get out of bed if you do not fall asleep immediately. Remember the goal is to associate your bed with falling asleep quickly! If you are in bed more than about 10 minutes without falling asleep and have not gotten up, you are not following this instruction.
  4. If you still cannot fall asleep, repeat step 3. Do this as often as is necessary throughout the night.
  5. Set your alarm and get up at the same time every morning irrespective of how much sleep you got during the night. This will help your body acquire a consistent sleep rhythm.
  6. Do not nap during the day.

Why it works

This method is based on the idea that we are like Pavlov’s drooling dog. We attach certain stimuli in the environment to certain thoughts and behaviours. Famously Pavlov’s dogs would start drooling when a bell rang, because they associated hearing the bell with getting food. Eventually the dogs would drool at the sound of the bell even when they didn’t get any food. Replace the bell with a bed and food with sleep and conceptually you’re there.

If we learn to do all kinds of things in bed that aren’t sleep, then when we do want to use it for sleep, it’s harder because of those other associations.

This is just as true of thoughts as it is of actions. It’s important to avoid watching TV in bed, but it’s also important to avoid lying in bed worrying about not being able to get to sleep. Because then you learn to associate bed with worry. Worse, you suffer anticipatory anxiety: anxiety about the anxiety you’ll feel when you are trying to get to sleep.

So, this therapy works by strengthening the association between bed and sleep and weakening the association between bed and everything else (apart from sex!).

Other treatments supported by the research are progressive muscle relaxation, which is exactly what it sounds like, and paradoxical intention. This latter technique involves stopping people trying so hard to get to sleep. The paradox being that when people stop trying so hard, they find it easier to fall asleep.

All this assumes you don’t live next door to a late night drummer and you’re not downing a double espresso before hitting the sack, but those sorts of things are pretty obvious. Everything else being equal, though, Stimulus Control Therapy seems the easiest for most people to implement.

Image credit: Meredith Farmer

No Comments