Review Article

A Comparative Study of Some Automatic Arabic Text Diacritization Systems

Table 4

Error rates of the HMM-based systems

SystemsDiacritic error rate: DER (%)Word error rate: WER (%)
Including ‘no diacritic’ classExcluding ‘no diacritic’ classIncluding ‘no diacritic’ classExcluding ‘no diacritic’ class
With case endingWithout case endingWith case endingWithout case endingWith case endingWithout case endingWith case endingWithout case ending

Mohamed Hadj Ameur et al.‘s work [21] (word-level HMM using Absolut discount smoothing technique)10.588.1312.369.4523.8412.2222.8811.80

Mohamed Hadj Ameur et al.‘s work [21] (word-level and letter-level HMMs using Absolute discount smoothing technique)9.747.1011.007.8423.9012.1922.8711.75

Mohamed Hadj Ameur et al.‘s work [21] (word-level HMM using Laplace smoothing technique)10.498.2012.229.5223.1012.0622.1611.67

Mohamed Hadj Ameur et al.‘s work [21] (word-level and letter-level HMMs using Laplace smoothing technique)9.657.1910.827.8623.1612.0222.1411.60