Hybrid Differential Evolution and Tabu Search for Parameter Tuning in Software Defect

Analyst’s Perception on the Use of AI-based Tools in the Software Development Life Cycle

“…In general, the challenges faced in most SDLC phases are related to the datasets used to test these AI techniques, such as limited dataset availability and low dataset quality (Chi et al 2023;Dhami et al 2021;Sonnekalb 2019). Apart from that, several challenges were also identified in various cases which may be related to certain phases in the SDLC such as chances of error (Sun et al 2022), high costs (Sonnekalb 2019), security concerns (Ozturk et al 2023;Sonnekalb 2019), inconsistent results (Ozturk et al 2023), and inaccurate result (Malhotra et al 2022) which is related to the implementation phase, there is concern about human error and the complexity of the model used (Chi et al 2023), which is related to the testing phase, as well as challenges related to the maintenance phase such as imbalance class in determining the model used to test the equipment (Tsoukalas et al 2022). Even though the study to examine the current condition of the use of AI in SDLC has been carried out, challenges, especially in the analysis phase related to the use of AI in SDLC, still remain unexplored.…”