EN
My tools

Keyword Extractor (RAKE)

Pull the key phrases out of any English text with the RAKE method: ranked phrases with scores and counts, plus a word frequency list. Runs in your browser.

Runs of more content words than this are skipped, as in the usual RAKE implementations.
Comma or space separated. For other languages, untick the built-in list and enter that language's common words here.
Top keyword
rich garden soil
Top score
7.5
Candidate phrases
16
distinct phrases after the filters
Words analysed
51
Keywords (one per line)

Scores follow RAKE: each word scores degree ÷ frequency (degree = the total length of the phrases it appears in), and a phrase scores the sum of its words. Longer phrases made of often-combined words rank highest.

1 run of more than 3 content words was skipped.

Key phrases ranked by RAKE score
RankPhraseScoreTimes found
1rich garden soil7.52
2garden soil5.171
3garden waste4.51
4kitchen scraps41
5compost bin3.51
6compost moist3.51
7rich2.331
8compost1.52
9mix11
10ideal11
11roses11
12months11
13turn11
14week11
15keep11
16wet11
Most frequent content words
WordCountShare of all words
compost59.8%
garden47.84%
rich35.88%
soil35.88%
bin23.92%
kitchen23.92%
scraps23.92%
ideal11.96%
keep11.96%
mix11.96%
moist11.96%
months11.96%
roses11.96%
turn11.96%
turns11.96%
waste11.96%
week11.96%
wet11.96%
My tools →
Advertisement

A keyword extractor finds the phrases that best sum up a piece of text — useful for tagging blog posts, checking what an article is really about before you optimise it, or summarising survey answers. This tool uses RAKE (Rapid Automatic Keyword Extraction), a published, transparent method: it splits the text at punctuation and common words, then scores each remaining phrase by how often its words appear and how often they appear together. Nothing is sent anywhere; the text is processed in your browser.

How to use this tool

  1. Paste the text — an article, a product description, a set of reviews.
  2. Choose the longest phrase to keep (3 words suits most pages).
  3. Raise the minimum count to keep only phrases that repeat.
  4. Add your own stopwords for words that should never be part of a keyword, then copy the list or read the tables.

Formula

score(word) = deg(word) ÷ freq(word); score(phrase) = Σ score(word)

Candidate phrases are runs of words between punctuation and stopwords. freq(word) counts how often the word appears in candidate phrases; deg(word) adds up the lengths of those phrases, so it is the frequency plus the number of co-occurring words. The method is from Rose, Engel, Cramer and Cowley (2010).

Example

In "Garden soil is rich in compost. Compost is a mix of kitchen scraps and garden waste. Rich garden soil is ideal for roses." the phrases are garden soil, rich, compost (twice), mix, kitchen scraps, garden waste, rich garden soil, ideal and roses. "garden" appears 3 times in phrases of total length 7, so it scores 7 ÷ 3 = 2.33; "soil" scores 5 ÷ 2 = 2.5 and "rich" 4 ÷ 2 = 2. The top phrase is rich garden soil with 2 + 2.33 + 2.5 = 6.83.

Frequently asked questions

How is this different from a keyword density checker?

A density checker counts how often given words or n-grams occur. RAKE finds the phrases themselves and ranks them by how strongly their words belong together, so it suggests multi-word keywords you did not think to look for. Use the keyword density checker to check a chosen keyword afterwards.

Does it work for other languages?

The method works for any language that separates words with spaces, but the built-in stopword list is English. Untick it and enter the common words of your language (for example articles, prepositions and pronouns) as extra stopwords.

Why are some long phrases missing?

Runs of more content words than the chosen length are skipped, which is how common RAKE implementations behave. Choose 4 or 5 words if your text has long technical terms.

Is my text stored?

No. The text stays in your browser and is not written into share links.

Sources

Last updated:

Cite or embed this tool
APA

kakaoCalc. (2026, November 8). Keyword Extractor (RAKE). https://kakao.mobile/tools/keyword-extractor/

MLA

“Keyword Extractor (RAKE).” kakaoCalc, 8 Nov. 2026, kakao.mobile/tools/keyword-extractor/.

Paste this code into your page. The widget runs in your visitors' browsers, shows no ads and keeps a link back to this page. Embedding terms

Advertisement