N-gram

N-Gram settings

• • N-gram size: choose 2-gram (default), 3-gram, 4-gram, 5-gram

  • Skip punctuations: hide punctuations

  • Output basis: Choose between token, POS tag, or lemma (only applicable when the corpus is POS tagged and lemmatised)

    • token : two red cars

    • POS tag : ART JJ NN

    • lemma : two red car

  • Click generate N-grams

Positional filter (optional)

• • You can choose how EACH gram is shown (similar to output basis but per gram). On 'basis' choose how you want to the gram to be shown

    • token POS POS ==> very JJ NN : very beautiful girl, very handsome man, very smart students

    • POS token token ==> ART red cars : some red cards, the red cars

  • You can use minus sign to show hide certain gram. for example, show all results but very

    • token POS POS ==> -very JJ NN : in this way, all results in gram one will show EXCEPT very

  • Note that this feature is only applicable if your corpus is POS tagged and lemmatised