Text frequency and character frequency analysis
Local statistics for Unicode characters, words, repeated lines, and basic text metrics; filter rules and export CSV with formula-injection protection.
How to use text frequency and character frequency analysis
The text is tokenized, counted, and exported only in the current browser; it will not be uploaded or saved.
Enter text
Paste UTF-8 content; the page accepts up to 1000000 characters and 50,000 lines, and further limits the encoded size to no more than 1 MiB.
Set statistics rules
Choose whether to ignore case, punctuation, and whitespace, and set the Top N count for 1–500.
View or export
View character frequency, word frequency, basic counts, and duplicate lines with line numbers, or export BOM and formula-injection-safe CSV.
Limitations and Notes
- Tokenization first uses Intl.Segmenter; when the browser does not support it, it falls back to the Unicode regex, and word boundaries may differ for Chinese, Japanese, and mixed text.
- The count of CJK only counts within the Unified Ideographs range and cannot be equated with the number of Chinese characters, words, or readable words in natural language.
- Items with the same frequency are sorted by first appearance; the ignore option changes normalized keys and values and cannot replace professional corpus analysis.
FAQ
Why might word frequency differ across browsers?
The browser-provided Intl.Segmenter implementation and fallback rules may use different word boundaries, so character frequency is usually more stable.
Why does the exported CSV add a prefix to the cells?
Content that starts with characters that trigger formulas will be prefixed safely to reduce the risk of it being executed as a formula in spreadsheet software.
Should repeated lines ignore whitespace?
Duplicate lines are handled according to the normalization options currently selected in the tool interface, and the corresponding original line numbers are retained for verification.