Annotating a big data-set for an Argument Mining task which is going to be processed by deep learning made me wondered if the presence or absence of punctuation(final dotes or a comma)affects the outcome of the process (it worth mentioning that some Argument Discourses Units- ADUs- are a paragraph long.
As you are annotating big data which is involving paragraphs, preprocessing of blogs would help in the removal of stopwords and nonwords from your text. Further, I would recommend using some suffix stripping algorithm so as to get only the stems of words and avoid duplication.
About your question, if we follow the above process, there will be a reduction in file size and surely it will help in minimizing the overall execution time of annotation.
I hope this helps.
- Badges
- Science topic