Skip to main content

Table 18 Performance of Jaccard coefficient in detecting near-duplicate documents (R8_NDD)

From: Pairwise document similarity measure based on present term set

 

k = 3

k = 4

k = 5

Precision

0.641

0.8065

0.8929

Recall

1

1

1

F-measure

0.7813

0.8928

0.9434