Painting the Historical Picture Through Data Mining
Text mining will be an essential component in visualizing the answers to the questions Contra will present. Since presedential debate transcripts are so widely available, we can use data mining to specifically find out how many times, The War on Drugs was mentioned by the candidates. This can give us a statistical basis to comparatively look at the funding that was pumped into the war on drugs in the subsequent presedential term. In this current election we see this domestic war scarcely being mentioned if at all by both candidates. I hypothesize that a correlation could emerge showing that the war continues to have an increase in funding, while it being less of a platform for campaigns to run on. Highlighting this information will open up further questions, such as; why? We could also illustrate on whether or not states support an increase, or decrease in budget within this war. This can be done by seeking out secondary sources in which different representatives give their personal opinion on the campaign.
Using data mining to assess the frequency with which candidates address the “War on Drugs” is, I think, just measuring the tip of the iceberg. My guess is that the secondary literature will provide confirmation of trends of this nature. However, there is room to cut deeper. You could look for patterns outside of debate transcripts and party platforms. You could possibility delve into the language of Americans across the country to assess the pervasiveness of the “War on Drugs.” It will be important for you to explain why the changes are occurring–the more nuanced a picture you have of the patterns, the better prepared you will be to do that.