Skip to main content

Table 18 Results of experiment III where the dataset is injected with plagiarism-simulated sentences and run against the entire corpus

From: An academic Arabic corpus for plagiarism detection: design, construction and experimentation

N-gram Segments

Segments in Test Dataset

Segments with Simulated Plagiarism

Segments Identified as Plagiarized

Reported Plagiarism Ratio

3

662

28

159

24.02%

4

672

19

57

8.48%

5

673

13

20

2.97%

6

673

8

11

1.63%

7

672

4

5

0.74%