Incorporating user preferences in search-based software engineering: A systematic mapping study

Abstract

ContextSearch-based algorithms have been successfully applied to solve software engineering problems in the field known as Search-based Software Engineering (SBSE). However, in practice, the user may reject the obtained solutions, since many characteristics of the problem cannot be mathematically modeled. To cope with this situation, preference-based algorithms have been investigated and raised interest in the SBSE field. ObjectiveTo identify the quantity and type of research on SBSE preference-based approaches and to contribute to this new research subject, named here Preference and Search-Based Software Engineering (PSBSE), MethodWe conducted a systematic mapping, following a research plan to locate, assess, extract and group the outcomes from relevant studies. ResultsFew software engineering activities have been addressed. The most used algorithms are evolutionary and single-objective. In most studies the preferences are provided interactively and, in many cases, the user preferences are incorporated in the fitness functions. We observe a lack of evaluation measures and works comparing existing approaches. ConclusionsThe use of preference-based algorithms in SBSE is an underexplored subject, and many research opportunities exist.