In 2012, a US academic named Dr. Martin Hilbert estimated that the average person was bombarded with 174 newspapers worth of information each day. He also estimated that each individual contributed 6 newspapers of data back into the world each day. Contrast this with 2.5 pages of data that each individual generated in 1986. It’s a lot of newspapers. It’s clear that we’ve stepped into a new era. The era of information and big data. And it’s only going to grow exponentially. In fact, it is estimated that by 2020 there will be 44 Zettabytes of data in our digital world. To put that into perspective, that’s 10Mb of information per grain of sand on Earth.
However, our faculties for processing this information haven’t changed as rapidly. We seem to live in a world with a constant information overload. A quick search on Google about a company or a person can reveal huge amounts of information. But gaining real insights from the information is still a challenge. What we need is an information roadmap. A way of altering our relationship with the data and connecting the dots. This is where the concept of data visualisation enters.
Data visualisation is the representation of data as graphic forms that the eye can comprehend. As humans our vision is exquisitely sensitive to variations in space, colour, shape and form. Our visual cortex has evolved over millions of years to detect and analyse visual patterns. This is important when you need to distinguish between a harmless animal and a predator trying to eat you. The entire structure of our brain is heavily biased towards analysing and understanding visual information. The aim of data visualisation is to utilise these evolved advantages and translate the language of the eye into the language of the brain. By doing this we can move beyond the data itself and start to understand the ideas and concepts hidden within the data. Our eyes give us the ability to unlock these hidden stories and patterns.
At MOBGEN:Lab, we are working on a data visualisation project which will allow users to interact with a dynamic visualisation of MOBGEN’s values and interests. Rather than handing visitors an info sheet about the company, the visualisation aims to allow visitors to visually and conceptually gain an overview of Mobgen. The vision is to project these data visualisations on the walls of the Lab creating an immersive and futuristic environment. A Microsoft Kinect can be used to allow visitors to physically interact with the data. Depth information from the camera will allow us to use the visitor’s motion in the space to control aspects of the visualisation.
Several experiments using Twitter data have also been undertaken in the lab. The Twitter API is a very accessible source of data. The Twitter servers can be queried for certain search words or hashtags and the server responds with a real time stream of relevant tweets. For example, if clients from an energy company visited Mobgen Lab, we could visualise tweets with relevant energy terms. This would allow the company to gain an insight into what twitter users were saying about their sector. The images below are some of the experiments into visualising a feed about a particular topic.
Another concept explored was using IBM’s Alchemy Language API to analyse the sentiment expressed in a tweet. The text in a tweet is analysed in the Alchemy language processing servers and a value is returned indicating how positive or negative the sentiment in the tweet is on a scale from -1 to +1. The graph below shows the results from searching for tweets related to Solar Energy. We can see how a higher proportion of the tweets analysed have a positive sentiment, which hopefully bodes well for the future of clean energy.
Ultimately the design goal of this project is to engage clients with MOBGEN:Lab, and make them feel inspired for their own projects and creative sessions. We want to show data in a creative way and inspire visitors with new insights into their own sector. We want to make the LabSpace an immersive and dynamic environment that hints at future possibilities that big data can offer us.