You want to see a brand new movie that opened last week, but your not sure if it’ll be good or not. Why not search the web for reviews? You have tons of websites to choose from: Rotten Tomatoes, iMDB, The Onion A.V. Club, etc... But rather than going through the labor of reading every one of those sites, you’d like to browse through a collection of the most viewed positive and negative reviews of the movie gathered from all over the web. Or perhaps you’d like to find out if most people on the internet thought highly of the movie based on a compiled statistical report of what has been written about the movie on blogs, online newspapers, or forums.
With the power of sentiment classification, this could all be possible in the near future. Computer systems with the ability to automatically evaluate the sentiment of a document will greatly benefit both consumers and businesses alike. Consumers will be able to search for information about products and people based on sentiment information – imagine being able to search for a baby carriage by asking a web search engine to find the carriage talked about on the web with the most praise and adulation. Businesses could have their sentiment classifying computer systems gather excerpts from the web of positive and negative user reviews of their products in order to improve their design processes. The applications for sentiment classification are limitless...