Table 2 Mann–Whitney U test result for the differences between AI tools and humans in interpreting metaphors in the two varieties of colloquial Arabic.

From: Metaphor interpretation in Jordanian Arabic, Emirati Arabic and Classical Arabic: artificial intelligence vs. humans

Group

N

Mean

Std. Deviation

Accuracy

Mann–Whitney U

Z

Sig. (2-tailed)

AI

4

8.75

3.40

43.8%

1.50

−3.150

0.002

Jordanian

29

17.52

2.18

87.6%

AI

4

15.50

1.29

77.5%

25.50

−1.232

0.218

Emirati

21

13.81

2.93

69.1%

Back to article page