ROUGE Score Calculator
Calculate the ROUGE score to evaluate text summarization quality against references.
About ROUGE Score Calculator
ROUGE (Recall-Oriented Understudy for Gisting Evaluation) measures summarization quality by counting n-gram overlap between a generated summary and reference summaries. Unlike BLEU which emphasizes precision, ROUGE focuses on recall — what fraction of reference content appears in the candidate.
Formula
ROUGE-N
Recall = overlap_ngrams / reference_ngrams; F1 = 2*P*R/(P+R)
- overlap_ngrams
- Number of matching n-grams between candidate and reference
Worked example
With the default values below, ROUGE Score Calculator returns ROUGE F1 of 0.6316. Change any input above and every figure here is recalculated in your browser.
Inputs used
- Overlapping N-gramsminimum 0
- 12
- Total Reference N-gramsminimum 1
- 20
- Total Candidate N-gramsminimum 1
- 18
Results
- ROUGE F1
- 0.6316
- ROUGE Recall
- 0.6000
- ROUGE Precision
- 0.6667
Step by step
- 1Recall: overlapping ÷ reference n-grams= 0.600000
- 2Precision: overlapping ÷ candidate n-grams= 0.666667
- 3F1: harmonic mean= 0.631579
Assumptions and limits
- The result is derived only from the 3 inputs listed above. Anything not among them is not part of the calculation.
- In particular, this calculator does not take inflation, tax, fees or charges or compounding frequency into account.
- Figures are rounded for display to 4 decimal places, so totals built from several rounded parts can differ slightly from the unrounded arithmetic.
How these calculators are built and tested is described on the Methodology page. If a figure looks wrong, please report it.
Frequently Asked Questions
What is the difference between ROUGE-1, ROUGE-2, and ROUGE-L?
ROUGE-1 counts unigram overlap, ROUGE-2 counts bigram overlap (captures fluency), and ROUGE-L uses longest common subsequence (captures sentence structure) without requiring consecutive matches.
When should I use ROUGE vs BLEU?
Use ROUGE for summarization tasks (where capturing reference content matters) and BLEU for translation tasks (where precision of generated text matters more).