Skip to main content

Table 5 The representation ability of the N-Grams for the class

From: Computing symmetrical strength of N-grams: a two pass filtering approach in automatic classification of text documents

N-Grams

Class C1

Class C2

Difference (D)

\(D^2\)

\(D^3\)

\(D^4\)

Nature of the N-Gram

\(t_{i}\)

2.3

2.25

0.05

0.0025

0.000125

0.00000625

Common

\(t_{j}\)

2.5

0.1

2.4

5.76

13.824

33.1776

Rare

\(t_{k}\)

2.5

0

2.5

6.25

15.625

39.0625

Very rare

\(t_{l}\)

0.05

0.01

0.04

0.0016

0.000064

0.00000256

Sparse