For the last couple of weeks I have been completely immersed reading Viktor Mayer-Schönberger’s, Big Data: A Revolution That Will Transform How We Live, Work, and Think. Whether you recognize it or not, we are now living in the era where big data has permeated every aspect of modern life, a fact Mayer-Schönberger’s book has opened my eyes to. Every time you use a phone or a computer, you’re not only dependent upon data, you’re also generating data. We seem to be inching closer to the science fiction of the movie The Matrix, where our lives could be fully represented by data. However, I digress, for this post isn’t about sci-fi, but reality.
“Big data” is definitely now on the media and corporate world’s radar. Big data analysis is offering new possibilities for decision making and general analysis in nearly any field that can be quantified. With the progression of data analyzing tools becoming more sophisticated, we’re finding new applications in fields as diverse as manufacturing, retail, and of course, the Internet of Things.
“Relationship” is the key word when it comes to big data analysis; in practice it’s a science more concerned about results rather than reason. Big data is being used as a testing pond: throw anything in and wait to see what it gives back to you. Some examples below:
This mapping is just one from a series of visualizations by Gnip, MapBox and Eric Fischer. A green dot indicates a Android user sending a tweet, a red dot indicates an iPhone, while a purple dot represents a Blackberry.
When I was at USC, I did this data mapping project shown above. The graphic is a visualization of the top 250 movies on IMDB and the Chinese equivalent website, Douban. I used the visualization tool Gephi to convert data into something showing the relationship between datasets. Here, the dataset includes selective basic information like name, year, country of origin, genre, and each film’s individual ranking. Each line represents a connection between two data points; one point can be dragged towards the other end.
Another visualization map shows the relationships between movies and variables such as their year, country of origin, whether the film ranked in the top 10 of IMDB or Douban, etc. Some conclusions can be made reviewing Douban’s top 250. For example, there are more Asian films listed, which is not a surprise. But other datasets reveal to be more interesting: it seems IMDBers enjoy old movies more than the Doubaners, while Doubaners like documentary films more than their IMDB counterparts.
Another really interesting finding from my treasure hunt about big data analysis is this infographic related to my own profession.
The data above was extracted from 928 job postings provided by 50 architecture firms. It reveals which software skills architecture students should be versed in. As you can see for yourself BIM (Building Information Modeling) has become vital, and software like Revit has surged in importance past traditional 2D CAD (although CAD does 3D,too).
This makes me wonder about how big data pertains to landscape design today and tomorrow. Landscape architecture design is a complicated process that involves data from so many different subjects, and it seems big data processing will become an increasingly vital tool for landscape architects and their projects. Geographic information system (GIS) might be the most well known tool currently, but other than that, what’s next?
If you find the topic of big data as fascinating as I do, I’d like to recommend the following fantastic big data visualizations sites: