≈Text Similarity (Levenshtein)
Calculate Levenshtein edit distance and percentage similarity between two strings. Useful for typo detection, similar-word lookup, search autocomplete, and data deduplication.
How to use
- 1Enter two texts.
- 2Edit distance and similarity (%) display instantly.
FAQ
Is it fast for long documents?+
Up to thousands of characters processes instantly. Very long texts (tens of thousands+) take longer.
What is Levenshtein?+
Minimum number of edits (insert, delete, replace) to transform one string into another. 'cat' → 'dog' takes 3 edits.
What does similarity % mean?+
100% = identical; 0% = totally different. 80%+ generally means very similar; below 50% is different.
Accurate for Korean?+
Yes. Calculations are syllable-level. For jamo-level comparison, pair with [Hangul Jamo Decomposition].
Where is it used?+
(1) Typo correction candidates, (2) DB duplicate row detection, (3) fuzzy search matching, (4) basic plagiarism checks.
Other similarity algorithms?+
Jaro-Winkler, cosine similarity, n-gram, etc., but Levenshtein is the most intuitive and common.