Productivity and Predictiveness of Grammatical Gender Assignment on Modern Standard Arabic's Nouns

Like many other Semitic languages,  Arabic language is known as having two genders to separate the masculine from the feminine, and distance itself from accepting the mistake in using them or confusing them, in order to preserve the integrity of the communication process and its continuation. Moreover, the linguistic research has indicated that there are difficulties in assigning grammatical gender in the Arabic language, as it is the case in many languages, which is related to distinguishing between masculine and feminine, the other is related to the multiplicities in the structural forms of gender whether natural or grammatical. Thus, the topic of masculinization and feminization is one of the most obscure topics that refuses to lead to an analytical principle. In addition, this paper is an attempt trying to present a proposed mechanism based on the use of an automatic learning algorithm according to a semantic network to build relationships between nouns, through which the gender is specified in the nouns in Modern Standard Arabic. The results of the application of the proposed mechanism confirm the possibility of predicting the gender of the nouns in a somewhat large extent, as the gender of the nouns cannot be completely predicted. This confirms that the difficulty continues, and the researches and investigations need to remain open, for more visions and new grammatical methods, to solve this issue.