Try some sample problems
Computational Linguistics
How would a computer figure out what a sentence means?
I shot an elephant in my pajamas.
How he got in my pajamas I don’t know! (Groucho Marx)
(source: http://specgram.com/CLIII.4/08.phlogiston.cartoon.zhe.html)
Examples of Computational Linguistics
Translating between languages automatically:
Predicting what words will come next:
Find out what different things males/females talk about on Twitter:
http://www.tweetolife.com/gender/
65% of tweets that use the word mouse were by males (and 35% by females). The list of “gendered features” shows the words that most commonly appeared in tweets with the word mouse. Are males and females writing about the same type of mouse? How would a computer figure out what type of mouse is being referred to?
Check spelling:
Showing ads specific to a user’s interests:
Find out what different texts have in common based on what words are used. You can use Wordle to visualize this. Wordle counts the number of times a word appears in a text and makes more frequent words bigger. Here are the word frequency visualizations for Oscar acceptance speeches. How would the visualizations look different if the words were from a science textbook or a news magazine?
Spam detection: How did the computer guess that this email was probably spam?
