The ngram size parameter (n) in redundancy calculation seems not used

[rordi]

Thank your for the wonderful work provided in the MoreThanSentiments package.

The function parameter n of the Redundancy method seems not used towards the ngram size, as the ngram size is hardcoded to 10 in the following line:

morethansentiments/src/MoreThanSentiments.py

Line 203 in ebb2837

ngram[i][j] = list(nltk.ngrams(input_data[i][j].split(),10))

As the original publication (Cazier and Pfeiffer, 2015) was based on rather long documents, the 10-gram was probably ok in that context. When dealing with shorter documents, it would be useful for users to be able to work with a smaller ngram size.

[jinhangjiang]

@rordi you are right. We will fix it soon. Thank you for pointing that out! Will get back to you on this thread once the bug is fixed.