Detecting two languages through knime or VBA or excel or spreadsheet

[Ranjit]

hi team, I have a query here. I have two different languages in one cell is it possible to find out from VBA or excel or spreadsheets or knime please let me know

Example: ASSEMBLAGE----French language

assembly–english language

Colur:Black,Assemblage— In these two languages are there how can I find out?

[Snowy]

So, just to be clear, you’re not only asking to detect if there are multiple languages, but also to determine what the languages are?

It may be helpful if you post a screenshot of exactly what your input looks like. This task seems like it could get complicated quickly as I’m sure there are many words that span across more than 1 language.

[bruno29a]

Hi @Ranjit , it is not clear what you are looking for.

“I have two different languages in one cell is it possible to find out from…” to find out what?

“In these two languages are there how can I find out?” find out what?

[Ranjit]

Hi what I am trying to asking is the in the first two cells i have one is English and another one having french that I can find out by using Google sheets formula as detect language but where as in the 3rd cell it contains both languages meana English and French but by using detect language formula I am not getting any error it is directly showing as English only even though 3rd cell having different language word.kindly let me if you understood or not

[SamirAbida]

Hi @Ranjit,

If I understand well your problem, you want to find the languages used in the comments. That means, that the comment can have more than one language (i.e. French and English for example).

One way to do this, could be to use a ngram model and a language detection. You can use NLTK for ngram (utils module) : NLTK :: nltk.util, that you can use like that:

from nltk.util import ngrams

n = 1 # can be changed by 2 for bigram, 3 fro trigram etc...
sentence = 'Some string i want to ngram.'
unigrams = ngrams(sentence.split(), n)

for item in unigrams:
    print(item)

[Out]
('Some',)
('string',)
('i',)
('want',)
('to',)
('ngram')

And you can use the library of your choice to detect the lang. Find here a great list of : string - Python: How to determine the language? - Stack Overflow

By the way, you can use Textblob for both functions (ngram and lang detection).

Br,

Samir