Selecting a standard set of attributes for cost estimation of software projects

The aim of the software engineering is to enhance projects that produce the needed results within limited schedule and budget. So that, software effort estimation becomes a valuable manner since it limits the problems of overestimate and underestimate for the software. Software cost estimation is the process of predicting the effort required to develop a software system. There are many estimation models over the last decade, and in this paper, we use six public cost estimation data sets that we obtained from promise repository. We perform regression analysis over these data sets and perform a feature selection in order to get the most effective attribute to the effort. Finally, we analyze and compare the results obtained from each data set to build a framework for the standard set of metrics that we suggest each cost estimation data set must contain.