Classified Arabic Documents Using Semi-Supervised Technique

In this work, we test the performance of the Naïve Bayes classifier in the categorization of Arabic text. Arabic is rich and unique in its own way and has its own distinct features. The issues and characteristics of Arabic language are addressed in our study and the classifier was modified and regulates to fit the needs of the language. a vector or word and their frequencies method is used to represent each document. We trained our classifier using both techniques supervised and semi-supervised in an attempt to compare between them and see if the classification accuracy will improve as a result of using the technique of semi-supervised. Many various experiments were performed, and the thoroughness of the classifier was measured using recall, precision, fallout and error. The outcomes illustrates that the semi-supervised learning can significantly enhance the classification accuracy of Arabic text.

@article{Khatatneh2016,
  title     = {Classified Arabic Documents Using Semi-Supervised Technique},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {7},
  number    = {5},
  year      = {2016},
  publisher = {The Science and Information Organization},
  author    = {Dr. Khalaf Khatatneh},
  doi       = {10.14569/IJACSA.2016.070503},
  url       = {https://doi.org/10.14569/IJACSA.2016.070503}
}