. . . . . .
This paper aims to provide a methodology for the stages that can be followed to build a Standard Arabic Test Corpus. The aim is to use it to automatically Word Sense Disambiguation in the texts of the classical Arabic language, especially since there is hardly a corpus that researchers can use to test their statistical linguistic models. Even there is no unified list of ambiguous words that can be subject to testing; which leads experts to a state of scientific uncertainty about the best proposed models for Arabic Word Sense Disambiguation automatically. The research followed a proposed approach, starting from defining the ambiguous words used in all ages, and passing through the collection and preparation of a major corpus, and ending with its classification in preparation for extracting a test set that represents the major corpus as much as possible.
. . . . .