Used for comparing the similarity of two samples. Also known as F1 score or Dice similarity coefficient. Majorly known as Sørensen–Dice index.

Dice Index is calculated using the formula:

Where X and Y are sets of features for each of the concepts and | | is cardinality of a set.

Example:
X = {1, 3, 5, 8}
|X| = 4

Y = {3, 7, 8}
|Y| = 3

X ∩Y = { 3, 8 }
|X ∩Y| = 2

Using the formula:
Similarity = 2 * 2 / (4 + 3) = 4 / 7 = 0.5714

Applications:

• Computer lexicography for measuring the lexical association score of two given words
• Image segmentation
• Comparing algorithm output against reference masks in medical applications