Size estimation through neural network for a common class of web-based systems

Abstract

This project deals with the analysis on the open source PhP systems size and through analysis on a number of open source software, a list of data based on the software’s database system were collected. These data significantly representing the characteristics of the software are “entities”, “relationships” and “attributes” found in the databases. And from these data, a backpropagation neural network was applied based on the method proposed in the conceptual model by taking entities, relationships and average attributes as the input, the lines of source code of a given software may be estimated.