EECS 498 Final Project, extracting emotion from the web.

{mawilkin, mkriete, parisedj, pdubslax} @ umich.edu

Human analysis of content takes a long time, and its difficult to extract the emotion from a set of web texts. We seek to take advantage of aggregated, user created web content to determine the general feeling of a given topic on the web.

Our web-based opinion engine operates as a web server that takes a query as input. We then use the Alchemy API to extract the primary subject of the query. With this subject, we perform a searches on bing and twitter to gather a real-time result set, which we then analyze.

• Requests -- http://docs.python-requests.org/
• Beautiful Soup 4 -- http://www.crummy.com/software/BeautifulSoup/

Install dependencies

python server.py -- This starts the server on port 3000

browser to http://localhost:3000/input.html

Input a query.