Companies can learn a lot about customer experiences by monitoring the social media web site Twitter. The file AirlineTweets contains a sample of 36 tweets of an airline’s customers. Normalize the terms by using stemming and generate binary term-document matrix.
a. What are the five most common terms occurring in these tweets? How often does each term appear?
b. Apply hierarchical clustering using complete linkage to yield three clusters on the binary term-document matrix using the tokens agent, attend, bag, damag, and rude as variables. How many documents are in each cluster? Give a description of each cluster.
c. How could management use the results obtained in part (b)?