Explainable Software Analytics

TL;DR

This paper emphasizes the importance of explainability in software analytics models to enhance trust and practical utility, proposing a research roadmap integrating social science and AI.

Contribution

It highlights explainability as a crucial aspect for software analytics models and outlines a research roadmap to develop more interpretable solutions.

Findings

Explainability increases trust in software analytics.

Current complex models lack sufficient interpretability.

A research roadmap integrating social science and AI is proposed.

Abstract

Software analytics has been the subject of considerable recent attention but is yet to receive significant industry traction. One of the key reasons is that software practitioners are reluctant to trust predictions produced by the analytics machinery without understanding the rationale for those predictions. While complex models such as deep learning and ensemble methods improve predictive performance, they have limited explainability. In this paper, we argue that making software analytics models explainable to software practitioners is as \emph{important} as achieving accurate predictions. Explainability should therefore be a key measure for evaluating software analytics models. We envision that explainability will be a key driver for developing software analytics models that are useful in practice. We outline a research roadmap for this space, building on social science, explainable artificial intelligence and software engineering.