Part 3. Case study 3: Tracking disease outbreaks using news headlines

 

Problem statement

Congratulations! You have just been hired by the American Institute of Health. The Institute monitors disease epidemics in both foreign and domestic lands. A critical component of the monitoring process is analyzing published news data. Each day, the Institute receives hundreds of news headlines describing disease outbreaks in various locations. The news headlines are too numerous to be analyzed by hand.

Your first assignment is as follows: You will process the daily quota of news headlines and extract locations that are mentioned You will then cluster the headlines based on their geographic distribution. Finally, you will review the largest clusters within and outside the United States. Any interesting findings should be reported to your immediate superior.

Dataset description

The file headlines.txt contains the hundreds of headlines that you must analyze. Each headline appears on a separate line in the file.

Overview

To address the problem at hand, we need to know how to do the following:

  • Cluster datasets using multiple techniques and distance measures.
  • Measure distances between locations on a spherical globe.
  • Visualize locations on a map.
  • Extract location coordinates from headline text.