Multi-Objective Search-Based Software Engineering

Many software engineering problems can be formulated as a computational search problem. In such problems, the search space is typically too large to be explored exhaustively. This leads to a rapidly growing research topic - search-based software engineering (SBSE) where computational search problems can be tackled with metaheuristics, e.g., evolutionary algorithms. When there is more than one objective/criterion to consider in the search, then the problem is referred to as multi-objective SBSE.

  • Pareto search vs Weighted search in multi-objective SBSE

      • [TOSEM22] Weighted search can be less promising than Pareto search even when clear preferences of the decision maker are available. [Read More]

  • Quality Evaluation in Multi-objective SBSE

      • [TSE22, ICSE-NIER18] A critical review and methodological guidance as to how to select and use evaluation methods in different Pareto-based SBSE scenarios. [Read More]