Software effort estimation: Harmonizing algorithms and domain knowledge in an integrated data mining approach

Abstract

Software development effort estimation is important for quality management in the software development industry, yet its automation still remains a challenging issue. Applying machine learning algorithms alone often can not achieve satisfactory results. In this paper, we present an integrated data mining framework that incorporates domain knowledge into a series of data analysis and modeling processes, including visualization, feature selection, and model validation. An empirical study on the software effort estimation problem using a benchmark dataset shows the effectiveness of the proposed approach.