Google Correlate 

  Google Correlate  is a new services from Google that allows you analyze temporal or spatial correlations between search terms. Could be incredibly interesting data in here for various fields. The methodology from the  whitepaper  gives an insight as to how Google does these correlations at scale: 

 
   In our Approximate Nearest Neighbor (ANN) system, we achieve a good balance of precision and speed by using a two-pass hash-based system. In the first pass, we compute an approximate distance from the target series to a hash of each series in our database. In the second pass, we compute the exact distance function on the top results returned from the first pass. 
 

 I tried my hand with  &#8220;Barack Obama&#8221; , where most of the action comes from the South, the Rust Belt, and the Eastern Seaboard:  

   

 Compare that to map for the search phrase  &#8220;Barack Hussein Obama&#8221; : 

   

 Here you see a much more distinct Appalachian pattern emerging. This map looks similar to others that highlight where John McCain in 2008 did better than George Bush in 2004. The kind of search terms with high spatial correlation with the two are also fascinating. For &#8220;Barack Obama&#8221; you see mostly references to African Americans or African American culture: 

   

 For &#8220;Barack Hussein Obama&#8221;, there is quite the hodgepodge, with references to &#8220;obama koran&#8221; and &#8220;obama the antichrist&#8221;: 

   

 Obviously, we wouldn&#8217;t want to read too much into these comparisons due to the ecological inference problems, but there is a lot to explore here.