Joshua Stevens of Pennsylvania State University presented the computer science-heavy "Accommodating Big Data in Visual Analytics: Pairing Computation with Cognition" at the 2013 Association of American Geographers conference.
His study shows how computer science can help understand big data in order to aid geospatial/geographic research.
While his presentation is not online he posted online an example of big data in visual analytics for geographic research.
His abstract is below
"Researchers in a variety of disciplines face enormous challenges related to the acquisition, analysis, and visualization of big data. This is especially true when the data contain both spatial and temporal attributes and when analytical procedures rely simultaneously on system computation and human cognition. For this reason, GIScientists are perhaps among those best prepared to make significant contributions to the emerging field of big data analytics. Intuitive and interactive visualizations that display rapidly changing, diverse spatiotemporal data will play a vital role in this process by improving the analytical and communicative stages of projects that make use of big data. The advent of big data brought along with it new challenges related to the velocity, complexity, and enormity of information and the pairing of many large, heterogeneous data sources. The challenges of visual analytics with big, spatiotemporal data are complicated further when maps only compose a portion of a visualization system consisting of animated graphs, tables, and other views that are employed alongside the computational processes. Although great strides have been made in developing tools to compute such data, questions related to the perception and cognition of big data faced by users of these tools remain unasked and unanswered. In light of these challenges, we make the case for considering cognition in the era of big data and introduce some of the efforts underway at Penn State related to GIScience and the dawn of big data visual analytics."