I am working on textual analysis of reports in English language. The documents are official records. I requires the sentence level segmentation of text. One way is through regular expression. Other way is through classifier trained on tagged data-set and then predict whether a period is sentence boundary or not? But for that i will require tagged data-set. Is there any sentence boundary tagged data-set freely available?
Some Corpora in the NLTK have a method
sents() that returns each sentence as a list of words, for example the Brown corpus:
import nltk nltk.download() # this opens a GUI to download all corpora needed from nltk.corpus import brown sentences = brown.sents()