Indonesian Journal of Electrical Engineering and Computer Science Vol. 22, No. 2, May 2021, pp. 961~967 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v22.i2.pp961-967  961 Journal homepage: http://ijeecs.iaescore.com Encapsulation of semantic description with syntactic components for the Arabic language Adel Al-Janabi, Ehsan Ali Al-Zubaidi, Radhwan Hussein Abdulzhraa Al Sagheer University of Kufa, Iraq Article Info ABSTRACT Article history: Received Jan 2, 2021 Revised Mar 5, 2021 Accepted Mar 12, 2021 The work presents new theoretical equipment for the representation of natural languages (NL) in computers. Linguistics: morphology, semantics, and syntax are also presented as components of subtle computer science that form. A structure and an integrated data system. The presented useful theory of language is a new method to learn the language by separating the fields of semantics and syntax. Keywords: Graph Lexical group Morphological features Natural language processing Semantics Syntax This is an open access article under the CC BY-SA license. Corresponding Author: Ehsan Ali Al-Zubaidi Department of Envorimental Planning University of Kufa, Iraq Email: ihsana.kareem@uokufa.edu.iq 1. INTRODUCTION In natural language, two autonomous phenomena are intertwined in NL: Discrete (grammar) and analog (semantics). Grammar (morphology and syntax) can be described in the language of mathematics [1]. Morphological the forms of words allow them to be distinguished and attributed they and their combinations have different values that they are recorded in the semantic dictionary [2]. In this case, this article shows you how to do this. The morphology model contains rules construction of word forms; word form of representation on the word in the original form and its morphological with parameters made up of decimal digits. We have the same morphological parameters forms a lexical group [3]. The model of NL syntax is based on the following assumptions are made: Mental sentences, that of the n members of the sentence you can always make up (n − 1) phrases (syntactically related pairs of word forms), such that the members of the sentence form a connected graph; in particular, combining simple sentences forms the syntactic graph Snt = (X, Y) from sets of X (word forms) and Y (word combinations) [4]. The degree of connectedness of two-word forms is determined by their morphological forms (parameters). That million of phrases, the Arabic language grammar, as well as they are divided into three hundred syntactic relations. Direct products of lexical groups, which makes it easier to represent the graph Snt in-memory computer [5]. The proposed network grammar model Snt opens up the following opportunities in the study of NL and its implementation on a computer: a) Root trees from the Snt model generate simple tree sentences. Bypass pre-location-tree assigns it a sentence sequence; correctness of the feedback a technique known as parsing offers depends on the extent to which the principles of fragmentation are observed. Between these sentences, the segments (tops of branches) included in the sentence tree fragments in the corresponding field sentence-sequence [6].