Data Science

What are n-grams?

- insted of dividing a document by each word, a document can be divided by n-grams
- n-grams meaning the n neighbors of a word
 
document: "The fox jumps over the fence"
3-grams: { The_fox_jumps, fox_jumps_over, jumps_over_the, over_the_fence}
 
- n-grams keeps part of the context the words occur in
 
-TFIDF can be calculated in the same way for n-grams as they are calculated for words

Diskussion