Spatial Data: Pattern and Process

The topic for this post comes from an article I've just read titled; Data Science and Prediction. In the article Vassant Dhar provides a brief overview of data science and new thinking about the relationship between theory and evidence. Dhar states the idea nicely saying that data science is different because it rests on the premise that the data can reveal the interesting questions. Traditional data analysis works the other way around starting with theory and looking to data for evidence to support or dispute the theory. Data science look for patterns in data and then ask if there are interesting questions those patterns might answer. This approach is not entirely new. The history of science is full of stories where an unexpected pattern in data led to new discoveries. What is new is that we now have a lot more data to analyze. Anyway, it's a good article, well worth reading.

The broad topics discussed in the article; data science and "big data" are of great interest to me. In fact, these topics are the central themes of my other blog. That's right, I have a second blog and it is called Pattern and Process.

Pattern and Process can into being first and it is intended for a different audience. Still, if you are reading this you might find it interesting. It's all about the technical issues and concerns that go along with research methods designed to find patterns in data; especially data that has a geographic or spatial context.

While I'm on the topic of influences and resources, I've also been reading In Search of Swampland by Ralph Tiner. This book provides a surprisingly readable introduction to the hydrology and biology of swamps (let's be honest, it's not the most exciting of topics). Central to the discussion is that swamps come in many

different varieties. This led the broad use of the term wetland to categorize lands where the presence of water changes both the form and the function of the place. This would be one of those rather dry academic subjects if it weren't for the regulatory protection of wetlands by governments at various levels. I'm sure I'll return to this topic at some point, there are marked wetlands in various parts of the Hemlock Forest, but for now the book is highly recommended.

One more book that will provide insight into the topic(s) of this blog is Forest Forensics, A Field Guide to Reading the Forested Landscape by Tom Wessels. Read this book and a walk through a forest in the northeastern United States will never be the same. Many forested areas in the northeastern U.S. were cleared land in the middle of the 1800s. This book helps you recognize and interpret clues that can tell the history of of how the land was used in the past and when that use was abandoned.





References:
Dhar Vassant, Data Science and Prediction, Communications of the ACM, 12/2013, Vol. 56 No. 12

Pattern and Process ( http://dataliterate.blogspot.com/ )

Tiner, Ralph; In Search of Swampland, Second Edition; Rutgers University Press, 2005

Wessels, Tom; Forest Forensics, A Field Guide to Reading Forested Landscapes, Countryman Press; September 20, 2010