Blog Post for Oct14

Our group chose to analyze assorted types of text and data because the program we were assigned to use, ManyEyes is efficient when visualizing all kinds of data. We used everything from charts and graphs to display things such as the 2008 demographics of Hawaii and crime rates in developing countries while we used word trees and word frequency applications for written texts.  


We used the word-tree application to analyze a text about Denmark.  Every time we would enter a word like "economic", you would return sentences like "economic... downturn in Denmark will diminish industrial production".  The word tree seems useful in determining if a topic of interest lies deep in a text.  For example, if there's a journal about Canada and its Climate and you want to see if they will be referencing climate change or the actual Canadian climate more, you could enter "climate".  Results would either be "climate... change is something Canadians must reverse" or "climate... in northern Manitoba seems to vary slightly through the spring months". After using the word tree as well as the other graphs and charts that were available, I determined that this application would be the most useful at text analysis, so long as you know what you're looking for.  I would say that among the data visualization tools, the clouds were pretty useful by taking data and grouping it into different sized and coloured clouds to represent different proportional values.


For my text analysis and data visualization exercise I plan on assessing W.W Denslow's "Three Bears".  A classic ferry tail of which I have familiarity will be perfect for this sort of analysis.  I think I will be using Wordle and ManyEyes in order to analyze the text due only slightly to familiarity, I feel ManyEyes has several tools which would be ideal for this kind of review.