In N-Grams How to export to xls only 2, 3, 4 N-Grams Results

jmphillips

Hello community: can you help me with this problem...... I want see in the end of the process in excel for example, only 2-Grams, or 3-Grams results. 

Thanks a lot !
Regards.

kayman

Use filter tokens. 
The logic of n-grams is that it adds underscores, so if you have 1 underscore you know it's a bigram, 2 is a trigram and so on. 
So a regex as below would filter out bigrams, but you can extend to add additional grams. 

^[^_]+_[^_]+$

Read as : start at beginning of word, take everything except an underscore, an underscore, and again everything except an underscore until the end of the word. So whatever matches must be a bigram. 

kayman

Use filter tokens. 
The logic of n-grams is that it adds underscores, so if you have 1 underscore you know it's a bigram, 2 is a trigram and so on. 
So a regex as below would filter out bigrams, but you can extend to add additional grams. 

^[^_]+_[^_]+$

Read as : start at beginning of word, take everything except an underscore, an underscore, and again everything except an underscore until the end of the word. So whatever matches must be a bigram. 

jmphillips

yo do it man !!! really thanks a lot for your help !!!

Regards