Engineering and Technology Journal,
2008, Volume 26, Issue 11, Pages 430-437
AbstractUsually, the natural language grammar is represented as a rules inside the
program of natural language processing (NLP), but with a language such as the
Arabic language that have a flexible syntax, using this representation will produce
a large number of these rules that lead to increase the search time required and to
increase the size of the program, so in this paper we proposed a new representation
method. We represent the Arabic language grammar as logical terms in the
knowledge base, and we use the B+tree to store these terms since it is suitable with
large data, so we reach the required syntactical term in a little time by using the
suitable searching method that b+tree provide it.
In this paper we choose the Augmented Transition Network(ATN) to represent
the language grammar since it is the most suitable parsing technique for a language
such as the Arabic language.
- Article View: 162
- PDF Download: 41