Closed-caption processing
èThe closed caption text is parsed and processed with a Natural Language Processing Tool
4
4Paragraphs, sentences and words are identified
4Each word is tagged with
lIts lemma , part-of speech, and morpho-syntactic properties:
– like number, person, form, tense, mode for verbs, positive, negative superlative for adjectives …
4Time codes are tagged like specific keywords
4
èResult is in XML