Keyword Density Analyser
TextEverything runs locally in your browser — nothing is uploaded
See which terms recur in a piece of text: every term with its count and share (of the total number of terms, not of characters), plus the total, the distinct count and the character count. You can filter common English and Chinese stop words, set a minimum length, and show only the top N. Chinese has no spaces, and this tool does not really segment it — it counts adjacent character pairs (bigrams), so a phrase becomes overlapping pairs, some of which are not real words; that is the method, not a bug. Counting single characters instead would be swamped by particles and carry almost no information. Everything runs in your browser; nothing is uploaded and nothing goes over the network.
Features
- Sorted by count, each term with its frequency and share
- Share is of the total number of terms, not of characters
- A built-in English and Chinese stop-word list, filtered in one tick
- Set a minimum length and show only the top N
- Optional bigrams (adjacent Chinese characters / adjacent English words)
- Language: automatic, Chinese, English, or mixed
- Reports the total, distinct count and character count
- Ignore case
- Everything runs locally in your browser; nothing is uploaded
How to use
- Paste the article to analyse (multi-line, mixed CJK/Latin is fine)
- Choose a language (automatic by default)
- Filter stop words, set a minimum length and top N, and switch bigrams on or off
- The result updates as you type; click Copy to take it
FAQ
- How accurate is the Chinese segmentation?
- This tool does not segment Chinese. Chinese has no spaces and real segmentation needs a dictionary and a statistical model; this uses the usual shortcut of adjacent character pairs (bigrams), so a phrase becomes overlapping pairs, some of which are not words — that is the method itself. It is chosen because single-character counts are swamped by particles and carry almost no information. For real segmentation, use a dedicated tool.
- What is the share a share of?
- Of the total number of terms (all tokens), not of characters. A term’s share is its count divided by the number of tokens. Using the character count as the denominator gives a percentage that is almost always a single digit, which is not what “density” means.
- What does stop-word filtering remove?
- A built-in list of common English and Chinese stop words: English like `the`, `and`, `is`; Chinese particles such as the possessive, perfective and copula markers. A Chinese bigram containing a stop character is dropped too, since such pairs are almost all noise. You can turn it off at any time.
- What does the bigram switch change?
- For Chinese, on (the default) counts adjacent character pairs, off falls back to single characters. For English, on additionally counts adjacent word pairs as `word1 word2`.
Related tools
Word Counter
TextCount Chinese characters and English words separately, alongside characters, lines, real UTF-8 bytes, reading and speaking time, and SMS segment counts.
Occurrence Counter
TextCount how often each character, word, line or given string appears, sorted by frequency with percentages. Spaces and breaks are labelled.
Text Inspector
TextFind invisible zero-width characters, BOMs and lookalike letters, and report normalization form, line endings and four character counts. All local.
Text Diff
TextCompare two blocks of text online and highlight added, removed and unchanged lines, so only the lines that changed are marked. Runs in your browser.
Text Cleaner
TextClean up a column pasted out of a log or spreadsheet: deduplicate, sort, drop blank lines, trim whitespace and strip line numbers, all in your browser.
Case Converter
TextConvert identifiers between fourteen naming styles. Words are split correctly first, so acronyms such as HTTPServer come out right, all in your browser.