Open AccessConstructing Expert System to Automatic Translation for Software development
Author(s) -
Safwan O. Hasoon,
Fatima Younis
Publication year - 2022
Publication title -
mağallaẗ al-kitāb li-l-ʿulūm al-ṣirfaẗ
Language(s) - English
Resource type - Journals
eISSN - 2617-8141
pISSN - 2617-1260
DOI - 10.32441/kjps.02.02.p16
Subject(s) - computer science , artificial neural network , knowledge base , expert system , artificial intelligence , source code , inference engine , coding (social sciences) , process (computing) , data mining , software , construct (python library) , inference , code (set theory) , set (abstract data type) , machine learning , programming language , statistics , mathematics
the development in computer fields, especially in the software engineering, emerged the need to construct intelligence tool for automatic translation from design phase to coding phase, for producing the source code from the algorithm model represented in pseudo code, and execute it depending on the constructing expert system which reduces the cost, time and errors that may occur during the translation process, which has been built the knowledge base, inference engine, and the user interface. The knowledge bases consist of the facts and the rules for the automatic transition. The results are compared with a set of neural networks, which are Back propagation neural network, Cascade-Forward network, and Radial Basis Function network. The results showed the superiority of the expert system in automatic transition process speed, as well as easy to add, delete or modify process for rules or data of the pseudo code compared with previously mentioned neural networks.