Research Article
Arabic Sign Language Recognition System for Alphabets Using Machine Learning Techniques
Table 10
Recognizing gestures letters accuracy by Holdout.
| No. of dataset | Classifiers and Holdout rate | Naïve-Bayesian | C4.5 | KNN | MLP | 66% | 75% | 66% | 75% | 66% | 75% | 66% | 75% |
| Dataset 1 | 84.11% | 83.04% | 83.65% | 78.00% | 90.58% | 88.38% | 86.90% | 84.19% | Dataset 2 | 89.50% | 88.90% | 94.05% | 89.57% | 97.67% | 96.81% | 95.72% | 95.28% | Dataset 3 | 97.73% | 98.24% | 97.74% | 95.05% | 99.62% | 99.02% | 98.71% | 98.59% |
|
|