Loading page
About this tool
Compare two texts as token sets and get the Jaccard index, distance, Dice, overlap and cosine coefficients.
Jaccard Similarity Calculator measures how much two texts overlap by turning each into a set of tokens and dividing the shared tokens by the combined tokens — the Jaccard index defined by Paul Jaccard in 1901. It also reports the Jaccard distance, the Dice–Sørensen coefficient, the Szymkiewicz–Simpson overlap coefficient and the Otsuka–Ochiai set cosine, and lets you tokenise by word, character, line or character n-gram shingles for near-duplicate detection.
Open Jaccard Similarity Calculator on AltFTool — it loads instantly in your browser.
Enter the values you already know.
Fine-tune the options to match your scenario.
Read the result and use it in your planning or reporting.
Jaccard, Dice, overlap and cosine all read the same intersection but normalise it differently, and seeing them together stops the wrong one being quoted.
Character n-grams catch reworded copy that word-level comparison scores as unrelated.
Shared, only-in-A and only-in-B lists make the number checkable instead of a black box.
Divide the number of tokens the two sets share by the number of distinct tokens across both. If set A has 4 words and set B has 4 words with 3 in common, the union is 4 + 4 − 3 = 5, so the Jaccard index is 3/5 = 0.6. Jaccard distance is simply 1 minus that, or 0.4.
Dice divides twice the intersection by the sum of the two set sizes, so it weights the shared elements more heavily and always returns a value at or above Jaccard. The two are exact transforms of each other: Dice = 2J/(1+J). A Jaccard of 0.6 is always a Dice of 0.75.
It depends entirely on the tokenisation, so there is no universal threshold. Near-duplicate detection on 3-character shingles typically flags documents above roughly 0.8, while topical similarity on word sets is often meaningful from 0.3 upward. Fix your tokenisation first, then calibrate a threshold on examples you have already judged.
No. Jaccard operates on sets, so a word appearing once and a word appearing fifty times contribute identically. If repetition matters, use cosine similarity over term-frequency or TF-IDF vectors, which is a different measure from the set cosine shown here.
Add the Jaccard Similarity Calculatorwidget to your blog or website — free, responsive, no signup. Just keep the “Widget by AltFTool” credit link visible.
<iframe src="https://www.altftool.com/embed/widget/jaccard-similarity-calculator"
title="Jaccard Similarity Calculator — free AltFTool widget"
width="100%" height="640" style="border:0;border-radius:12px;overflow:hidden"
loading="lazy" referrerpolicy="no-referrer-when-downgrade"></iframe>
<p style="font-size:12px;margin:4px 0 0">Widget by <a href="https://www.altftool.com/tools/all/jaccard-similarity-calculator?utm_source=embed&utm_medium=widget">AltFTool — free online tools</a></p>