Search

This paper proposes an effective wrapper approach by integrating the Levy flight and evolutionary selection operators into the Moth Flame Optimization (MFO) algorithm. The main purpose is to solve the Feature Selection (FS) problem in medical applications. FS is used as a preprocessing step in a data mining process to improve the performance of the classification system by eliminating irrelevant and redundant features from a dataset. This simplifies the generalization process and reduces the complexity of the generating models. Furthermore, it speeds up the learning process and reduces the cost for additional hardware resources. However, FS is a challenging NP-hard problem because the search space grows exponentially with an increase in the number of features. In this paper, the swarming behavior of the moths is utilized by using the MFO optimizer as a search strategy within a wrapper approach. The Levy flight operator is proposed to enhance the exploratory behavior of the MFO and mitigate the stagnation in local minima. Different selection mechanisms: random selection (RS), tournament selection (TS), and roulette wheel selection (RWS) methods, are investigated to decrease the bias of the MFO algorithm toward exploitation. These selection operators are proposed in the combination of the Levy flight in the form of four different FS methods LBMFO-R1, LBMFO-R2, LBMFO-TS, and LBMFO-RWS. The proposed methods are validated using 23 medical data sets from well-regarded data repositories. The comprehensive results and various comparisons reveal that the Levy flight and selection operators have a great positive impact on the performance of the MFO. They enhance the exploration, convergence trends, and diversity of solutions. [ABSTRACT FROM AUTHOR]

Software defect prediction (SDP) is crucial in the early stages of defect-free software development before testing operations take place. Effective SDP can help test managers locate defects and defect-prone software modules. This facilitates the allocation of limited software quality assurance resources optimally and economically. Feature selection (FS) is a complicated problem with a polynomial time complexity. For a dataset with N features, the complete search space has 2 N feature subsets, which means that the algorithm needs an exponential running time to traverse all these feature subsets. Swarm intelligence algorithms have shown impressive performance in mitigating the FS problem and reducing the running time. The moth flame optimization (MFO) algorithm is a well-known swarm intelligence algorithm that has been used widely and proven its capability in solving various optimization problems. An efficient binary variant of MFO (BMFO) is proposed in this paper by using the island BMFO (IsBMFO) model. IsBMFO divides the solutions in the population into a set of sub-populations named islands. Each island is treated independently using a variant of BMFO. To increase the diversification capability of the algorithm, a migration step is performed after a specific number of iterations to exchange the solutions between islands. Twenty-one public software datasets are used for evaluating the proposed method. The results of the experiments show that FS using IsBMFO improves the classification results. IsBMFO followed by support vector machine (SVM) classification is the best model for the SDP problem over other compared models, with an average G-mean of 78%. [ABSTRACT FROM AUTHOR]