Analyzing Tweets to Draw Inferences from the Domestic Airline Sector

I wanted to choose a problem which directly affects the bottom-line of a company. There are two ways to do this: increasing revenues or decreasing costs. Now, it is very difficult to decrease costs in a market such as the airline industry where competition is steep and price for labor and suppliers is fairly constant. One of the biggest differentiators in this space (as in most others) is customer service.

My capstone project included leveraging twitter data to conduct a sentiment analysis on domestic airlines. Using traditional regression and classification methods, I was able to create a “Customer Improvement Proposal” and give an insight on how to improve the overall customer experience and where resources should be appropriately allocated.