Compare vocabulary between two texts

Compare the distinct meaningful words in two English texts with Jaccard set similarity.

freeworks offlinenothing uploaded
ToolWord Overlap Diff
Input
Output
—
Put this on your own site

The frame below runs the same code as this page, in the reader's own browser. Nothing is sent to us, and nothing is sent to you.

Pick a dark background and the text and panels follow it, so the frame stays readable on a dark page.

Preview

How it works

Both sides are lowercased, ASCII punctuation becomes spaces, tokens under three characters and a fixed English stop list are discarded, and each remainder becomes a set. Similarity is the size of the shared set divided by the size of the combined set; repeated occurrences and word order have no effect.

Formula
Jaccard similarity = shared distinct words ÷ all distinct words across both texts
  • The result lists at most 20 shared and side-specific tokens so a long document cannot dominate the page.
  • Labels at 20%, 50%, and 80% are editorial bands, not confidence or plagiarism thresholds.

Worked example

Compare two paragraphs
Find word overlap between two related texts
Input
											Input: JavaScript is a popular programming language for web development.
Second: Python is a popular programming language for data science.
Output
												Similarity: 33
Common words: popular, programming, language
Unique to first: javascript, web, development
Unique to second: python, data, science
Analysis: Texts have some overlap but are mostly different.

When to use this

Editors compare vocabulary retained between revisions, search analysts inspect query-to-copy overlap, and classrooms demonstrate Jaccard similarity on word sets.

Edge cases

  • dog repeated four times contributes the same one set member as dog written once.
  • not is removed by the English stop list, so an affirmative and its negation can appear closer than their meanings.
  • café is split by an ASCII word pattern, making multilingual comparison incomplete.