Cost estimation of web applications through knowledge elicitation

Abstract

Objective - The objective of this paper is detail the use of tacit knowledge elicited from domain experts in the domain of Web effort estimation to build an expert-based Web effort model for a medium-size Web company In Auckland (New Zealand). Method - A single-company Web effort estimation model was built using Bayesian Networks (BN), using knowledge solely elicited from two domain experts who were experienced Web project managers. The model was validated using data from eleven past finished Web projects. Results - The BN model has to date been successfully used to estimate effort for numerous Web projects developed by this Company. Conclusions - Our results suggest that, at least for the Web Company that participated in the case study, the use of models that allow the representation of uncertainty, inherent in effort estimation, can outperform expert-based estimates. Thus far, another nine companies in New Zealand, and on in Brazil have also benefited from using Bayesian Networks, with very promising results. © 2012 Springer-Verlag.