What was Nate Silver’s key insight about the available data?
What were the technical challenges that he had to overcome?
View the two introductory videos. The first video presents several interesting examples of data science at work. The second video also goes into some detail about the source of the data and methodologies. The questions from the first video are a bit abstract and the questions from the second video are more direct.
Pick one video and answer questions. Find at least one article or website or blog post that discusses the related challenge or example. Include the link to the reference article or post, and write a one-paragraph short response to each question.
Here is the overall instruction which I forgot to put
Obama’s campaign outreach:
What different technologies were used and for what types of queries? Any surprises?
Hurricane Sandy:
What are the interesting data science issues discussed in this example?
Expression of emotions:
What data do we get from Google and WordNet to solve this problem?
What is an N-gram? What is normalization? Why are they important here?
Increase in fear in the 90’s: Do you agree with this conclusion? Is there any corroborating evidence?
What is the interesting conclusion from slide 10?
Video2:
Slide 11:
What does the PageRank algorithm do on the Web? Can it work on a citation graph?
Slide 12:
How can you cluster documents based on topics? What interesting result does this figure show?
Slide 13:
How was the graph constructed? What algorithms are mentioned?
Slide 14:
What data was re-purposed?
Google flu trends:
What can go wrong with a prediction?
Hyperglycemia:
What is the insight here? Are there other such examples?
Sample Solution