paragraph_bayesian.md

• code description: A Bayesian classifier is trained to automatically categorize the subtitles. The labeled subtitles are fed into the readtrain function, and the training set and test set are divided in line 30 to line 33 of the code. Finally, a Bayesian classifier named CLF is obtained.
• input: Labeled subtitles
• output: A Bayesian classifier that can categorize a subtitle into one of the types listed in Table 1 ***Notice: *** The training method is the same as that in the sentence-level classifier (see sentence_bayesian.py), except that in the latter, the labeled data are sentences.