Abstract:Ensemble learning has been widely used in various fields. Still, too many base classifiers will affect the classification time of the ensemble classifier under the big data environment, while reducing base classifiers will affect the classification accuracy of the ensemble classifier. Therefore, the multi-objective teaching-learning-based optimization (MO-TLBO) algorithm is used to carry out ensemble pruning of random forest (RF) to improve the classification accuracy and speed of RF. MO-TLBO algorithm aims at… Show more
“…Therefore, many traditional metaheuristics such as genetic algorithm (Lu et al, 2020), particle swarm optimization (Al-Sawwa & Ludwig, 2020), differential evolution (He et al, 2021), whale optimization (AlJame et al, 2020), sine cosine (Alfailakawi et al, 2021), teaching-learning-based optimization (Wan et al, 2021), and grey wolf optimizer (Jarray et al, 2022a) have been successfully parallelized on Spark environments showing considerable performance gains for large scale problems. However, AOA being a recently proposed metaheuristic has not been parallelized under such an environment yet.…”
Arithmetic optimization algorithm (AOA) is a recent population-based metaheuristic widely used for solving optimization problems. However, the emerging large-scale optimization problems pose a great challenge for AOA due to its prohibitive computational cost to traverse the huge solution space effectively. This article proposes a parallel Spark-AOA using Scala on Apache Spark computing platform. Spark-AOA leverages the intrinsic parallel nature of the population-based AOA and the native iterative in-memory computation support of Spark through resilient distributed datasets (RDD) to accelerate the optimization process. Spark-AOA divides the solutions population into several subpopulations that are distributed into multiple RDD partitions and manipulated concurrently. Simulation experiments on different benchmark functions with up to 1,000-dimension and three engineering design problems demonstrate that Spark-AOA outperforms considerably standard AOA and Spark-based implementations of two recent metaheuristics both in terms of run-time and solution quality.
“…Therefore, many traditional metaheuristics such as genetic algorithm (Lu et al, 2020), particle swarm optimization (Al-Sawwa & Ludwig, 2020), differential evolution (He et al, 2021), whale optimization (AlJame et al, 2020), sine cosine (Alfailakawi et al, 2021), teaching-learning-based optimization (Wan et al, 2021), and grey wolf optimizer (Jarray et al, 2022a) have been successfully parallelized on Spark environments showing considerable performance gains for large scale problems. However, AOA being a recently proposed metaheuristic has not been parallelized under such an environment yet.…”
Arithmetic optimization algorithm (AOA) is a recent population-based metaheuristic widely used for solving optimization problems. However, the emerging large-scale optimization problems pose a great challenge for AOA due to its prohibitive computational cost to traverse the huge solution space effectively. This article proposes a parallel Spark-AOA using Scala on Apache Spark computing platform. Spark-AOA leverages the intrinsic parallel nature of the population-based AOA and the native iterative in-memory computation support of Spark through resilient distributed datasets (RDD) to accelerate the optimization process. Spark-AOA divides the solutions population into several subpopulations that are distributed into multiple RDD partitions and manipulated concurrently. Simulation experiments on different benchmark functions with up to 1,000-dimension and three engineering design problems demonstrate that Spark-AOA outperforms considerably standard AOA and Spark-based implementations of two recent metaheuristics both in terms of run-time and solution quality.
“…Major power networks are also growing more sophisticated as the world's energy usage rises. However, the usage of alternative energy sources has rapidly increased as a result of rising worries about global pollution and the finite supply of fossil fuels [4][5][6]. Due to its capacity to lower pollution, renewable energy has recently gained popularity and currently provides a sizeable amount of the world's power [7][8].…”
In order to remove undesirable lower order harmonics from a cascaded H-bridge multilevel inverter and solve a constrained in design optimization issue with parameters involved in a changeable objective function for the given system, the TLBO approach is provided. To test the stability of the system, the recommended system output was designed to be supplied to various load drive systems. The entire efficiency of wind grid inverters has now decreased due to the low wind speed, making it unable to utilise wind energy effectively. To solve this issue, it is necessary to improve both the control method and the inverter side’s topological structure. A cascaded H-bridge multilevel inverter architecture is used to implement the entire system’s functionality, with the first stage circuit being built in MATLAB-Simulink.
Ensemble learning has become a cornerstone in various classification and regression tasks, leveraging its robust learning capacity across disciplines. However, the computational time and memory constraints associated with almost all-learners-based ensembles necessitate efficient approaches. Ensemble pruning, a crucial step, involves selecting a subset of base learners to address these limitations. This study underscores the significance of optimization-based methods in ensemble pruning, with a specific focus on metaheuristics as high-level problem-solving techniques. It reviews the intersection of ensemble learning and metaheuristics, specifically in the context of selective ensembles, marking a unique contribution in this direction of research. Through categorizing metaheuristic-based selective ensembles, identifying their frequently used algorithms and software programs, and highlighting their uses across diverse application domains, this research serves as a comprehensive resource for researchers and offers insights into recent developments and applications. Also, by addressing pivotal research gaps, the study identifies exploring selective ensemble techniques for cluster analysis, investigating cutting-edge metaheuristics and hybrid multi-class models, and optimizing ensemble size as well as hyper-parameters within metaheuristic iterations as prospective research directions. These directions offer a robust roadmap for advancing the understanding and application of metaheuristic-based selective ensembles.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.