Provably Optimal and Human-Competitive Results in SBSE for Spectrum Based Fault Localisation

摘要: Fault localisation uses so-called risk evaluation formulae to guide the process. For more than a decade, design and improvement of these has been conducted entirely manually through iterative publication in fault literature. However, recently we demonstrated that SBSE could be used automatically such by recasting this as problem for Genetic ProgrammingGP. In paper prove our GP produced four previously unknown globally optimal formulae. Though other human competitive results have reported literature, is first result, any application domain, which competitiveness formally proved. We also show some exhibit counter-intuitive characteristics, making them less likely found solely further effort.