Perspectives on Data Science for Software Engineering

Becoming Goldilocks

Privacy and data sharing in “just right” conditions

F. Peters    Lero - The Irish Software Research Centre, University of Limerick, Limerick, Ireland

Abstract

One of the facets of data science is reproducible reporting. To do this, data used for analysis must be shared, but this is not always possible due to business sensitivity and/or privacy concerns. In this chapter we share valuable lessons learned from years of data privacy research in the field of Cross Project Defect Prediction. These lessons can be considered when developing or using privacy algorithms for different types of software data.

Keywords

Data privacy; Data sharing; Cross-project defect prediction; Sensitive attribute disclosure; MORPH; Inverse Privacy ...