Computational Intelligence Applications For Software Engineering Problems Parma Nand by Parma Nand, Nitin Rakesh, Arun Prakash Agrawal, Vishal Jain, (eds.) 9781774910467, 9781774910474, 9781003283195, 1774910462, 1774910470, 1003283195, 2022026648 instant download after payment.

Size and complexity of software systems are increasing day by day, and so are the challenges associated with them. Before becoming obsolete, any software undergoes a number of stages during its lifecycle, such as requirements engineering, design, coding, testing, and maintenance to name a few. Each of these stages accommodates a number of costly and error-prone activities that are performed. Computational intelligence techniques can be applied to carry out these activities effectively and efficiently. Computational intelligence techniques are aimed at providing better and optimal solutions to real-world complex optimization problems in reasonable time limit. These are also closely related to artificial intelligence (AI) and incorporate heuristic as well as metaheuristic algorithms. These approaches have successfully been applied to solve real-world problems in various application domains such as healthcare, bioinformatics, civil engineering, computer networks, scheduling, software project planning, resource allocation, and forecasting, to name a few. These techniques have also attracted researchers in the software engineering domain and have been successful in prioritization of requirements, size and cost estimation of software to be developed, software defect prediction and reliability assessment, test case prioritization and vulnerability prediction, and many more. Size of solution space in such domains is very large, and intelligent behavior shown by computational intelligence techniques including evolutionary algorithms, machine learning algorithms, and metaheuristic algorithms find appropriate application of these approaches.
Machine learning approaches are, however, constrained by the availability of huge amount of data to extract knowledge and to build and train the model. A metaheuristic algorithm, on the other hand, is a high-level, iterative process. Exploration and exploitation are the two key characteristics of any metaheuristic algorithm that guides
…