确定 python 字符串中的字边界

Question

I have filepaths in the format of:

THISISSOMEMOVIE.mov

Is there some NLP library that can make very educated/statistical guesses about the word boundaries in a string? For example, the above should be parsed as:

THIS IS SOME MOVIE mov

Answer 1

I don't know of a library that does just that but you could use PyEnchant that tells you if a word belongs to the dictionary.

So here's the pseudo code of what I'd do:

 s = 0
 i = len(title) - 1
 check if the substring s-i is in the dictionary
    if not i = i - 1
    if yes then s becomes i+1, and i = len(title) - 1 again