Given that any words in E can be mapped to any words in F, it follows that there are x alignments that can be built between two sentences (E, F) of lengths l and m, respectively.
* *
m > = 1/e (1)
The inequality (1) establishes a sort of worst-case general bound, but unfortunately this bound turns out to have limited utility in our WSD application.
Go to main page for English language usage