Jaccard Similarity Calculator

Jaccard Similarity Calculator

How similar are two sets? Jaccard similarity answers that question by comparing what they share against everything either one contains.

J(A,B) = |A ∩ B| / |A ∪ B|

Worked Example

• Set A: apple, banana, cherry
• Set B: banana, cherry, date

Intersection = {banana, cherry} (size 2). Union = {apple, banana, cherry, date} (size 4).

Jaccard Similarity = 2/4 = 0.5

Reading the Score

A score of 1.0 means the sets are identical; a score of 0.0 means they share absolutely nothing. Values in between give a proportional sense of overlap that's independent of how large either set is - two sets with only 2 items in common feel very different from a shared-2-out-of-1000 overlap, and Jaccard similarity captures that distinction correctly.

Common Applications

Jaccard similarity is widely used in text and document comparison (treating documents as sets of words or shingles), collaborative filtering recommendation systems (as sets of user preferences or purchased items), and deduplication pipelines that need to flag near-identical records based on shared attributes.