What is the Jaccard similarity index?

The Jaccard index (also called Jaccard similarity coefficient) is the size of the intersection of two sets divided by the size of their union. A score of 1 means the sets are identical; 0 means they share no elements.

Is the comparison case-sensitive?

No. Tokens are normalized to lowercase before comparison, so "Apple" and "apple" are treated as the same element.

What separators can I use?

You can use commas, spaces, or any mix of both. The calculator splits on commas and whitespace, so "cat, dog bird" and "cat dog, bird" produce the same set.

No. Jaccard similarity is defined on sets, so the order of tokens has no effect on the score. Duplicate entries within a single set are also ignored.

What is the difference between Jaccard and cosine similarity?

Jaccard works on sets (presence/absence of elements), while cosine similarity works on vectors (it accounts for term frequency). For binary presence/absence scenarios, Jaccard is simpler and often more interpretable.

Jaccard Similarity Calculator

Name: Jaccard Similarity Calculator
Availability: InStock
Author: Nham Vu

Compute the Jaccard similarity index between two sets — paste comma- or space-separated tokens and instantly see intersection, union, and similarity score.

Set A

Comma- or space-separated tokens

Unique tokens: 0

Set B

Comma- or space-separated tokens

Unique tokens: 0

Jaccard Index

—

Intersection

—

tokens in both sets

Union

—

total unique tokens

Similarity —

Summary

Compute the Jaccard similarity index between two sets — paste comma- or space-separated tokens and instantly see intersection, union, and similarity score.

How it works

Paste or type tokens into Set A and Set B — separate them with commas or spaces.
The calculator splits each input into unique tokens, ignoring duplicates and case.
It computes the intersection (tokens in both sets) and the union (all unique tokens combined).
The Jaccard index is calculated as |A ∩ B| / |A ∪ B|.
Results update live as you type, showing the score, intersection, and union lists.

Use cases

Measure document or keyword overlap for plagiarism detection.
Compare product tag lists to find similar items in a catalog.
Evaluate recommendation system quality by checking predicted vs. actual item sets.
Assess gene set similarity in bioinformatics research.
Deduplicate datasets by identifying near-duplicate records.
Validate NLP tokenizer output by comparing expected vs. actual token sets.

Frequently Asked Questions

Last updated: 2026-07-22 · Reviewed by Nham Vu