I\'m trying to build a dataset consist of certain part of document. For example, the document format is like this:
According to A : Lorem Ipsum is si