So on the development set of 270 cases, we get 80% correct in 17 seconds (a rate of 16 Hz), and on the final test set we actually do a little better, 87% correct (but at only 14 Hz). Therefore I can declare victory: the program is doing better than 80% and better than 10 corrections/second.
model = collections.defaultdict( lambda : 1 )
A very senior Microsoft developer who moved to Google told me that Google works and thinks at a higher level of abstraction than Microsoft. "Google uses Bayesian filtering the way Microsoft uses the if statement," he said. That's true. Google also uses full-text-search-of-the-entire-Internet the way Microsoft uses little tables that list what error IDs correspond to which help text. Look at how Google does spell checking: it's not based on dictionaries; it's based on word usage statistics of the entire Internet, which is why Google knows how to correct my name, misspelled, and Microsoft Word doesn't.
« Older The Spirit of Truth | Stem cell research: Natural Born Killers? Newer »
This thread has been archived and is closed to new comments