3 documents
d0: "the cat sat"
d1: "the dog sat"
d2: "the cat ran"
vocab:
the, cat, sat,
dog, ran
count
BoW · d0
[1, 1, 1, 0, 0]
BoW · d1
[1, 0, 1, 1, 0]
BoW · d2
[1, 1, 0, 0, 1]
* IDF
down-weight
"the"
TF-IDF · d0
[0.0, 0.58, 0.58, 0.0, 0.0]
TF-IDF · d1
[0.0, 0.0, 0.58, 0.82, 0.0]
TF-IDF · d2
[0.0, 0.58, 0.0, 0.0, 0.82]
order discarded. rarity rewarded. sparse, interpretable, fast.