Chaotic vortex search algorithm: metaheuristic algorithm for feature selection

Gharehchopogh, Farhad Soleimanian; Maleki, Isa; Dizaji, Zahra Asheghi

doi:10.1007/s12065-021-00590-1

Cited by 98 publications

(42 citation statements)

References 88 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The simulation experiments of the proposed GSIWOA-RM and the benchmarked drift adaptive-deep reinforcement learning-based scheduling (DA-DRLS), 26 GWO algorithm-based resource management (GWOA-RM), 25 and SERA 28 approaches are conducted using Keras with the support of Theano python3 library that aids in optimizing the mathematical evaluations. [36][37][38] This implementation is carried out with the system configuration of Intel ® Core™ i5-6200U CPU @ 2.30 GHz, 8-GB RAM, and 2.40 GHz with 64-bit operating system. The emulation for portraying the real-time IoT scenario is attained with the arrival rates of service ranging between 0.1 and 1.0.…”

Section: Simulation Results and Discussionmentioning

confidence: 99%

Galactic swarm‐improved whale optimization algorithm‐based resource management in Internet of things

Karthick

Gomathi

2021

Int J Communication

View full text Add to dashboard Cite

Summary Internet of things (IoT) is the reliable alternative among the networking technologies for achieving high performance with improved potentialities of flexible adoptions, data exchanges, resource allocations, and system controls. The existing IoT suffers from the limitations of resource allocation ranging between complicated service provisioning environments and networking service quality mismatching. IoT environment needs to handle the resource allocation issue for attaining satisfactory degree of quality of experience (QoE) that maps multiple resources to gateways. This problem of mapping multiple resources to gateways belongs to the class of NP‐complete problem and can be ideally solved through intelligent metaheuristic algorithms. In this paper, galactic swarm‐improved whale optimization algorithm‐based resource management (GSIWOA‐RM) scheme is proposed for efficient mapping of multiple resources to gateways in IoT. It specifically utilizes galactic swarm optimization algorithm (GSOA) for establishing global control with inherited multiple adaptive cycles of exploitation and exploration during resource allocation. It further prevents the limitations of early convergence in the exploitation phase by utilizing the evolution‐based whale optimization algorithm (WOA) that aids in better balance between exploitation and exploration. The simulation results of the proposed GSIWOA‐RM scheme confirm a better throughput of 28.32% with minimized delay and energy consumptions of 19.24% and 21.82%, when compared to the baseline resource management schemes.

show abstract

Section: Simulation Results and Discussionmentioning

confidence: 99%

Galactic swarm‐improved whale optimization algorithm‐based resource management in Internet of things

Karthick

Gomathi

2021

Int J Communication

View full text Add to dashboard Cite

show abstract

“…Most papers discussed in the literature assessed the performance of their techniques based on a collection of wellknown instances with various scales (small, medium, and large) to check their stability in addition to their ability to find better results [73], [84], [117]- [121]. After observing those papers, we found that most employed instances were taken from the UCI repository.…”

Section: A Datasets Descriptionmentioning

confidence: 99%

“…Therefore, most of those instances in addition to others taken also from this repository with a number of up to 32 instances have been here employed to validate our proposed algorithm compared to some of the rival algorithms discussed later. As aforementioned that we employed those instances, as they have been widely used in the literature as an attempt to achieve a fair comparison within our experiments [29], [30], [33], [36], [73], [84], [117]- [121]. Generally, those employed instances are described in Table 2 to illustrate their characteristics such as the number of features (#F), number of classes (#C), and number of samples (#S).…”

Section: A Datasets Descriptionmentioning

confidence: 99%

“…There are much more metaheuristic algorithms tackled the FS problem, such as sine cosine optimization algorithm (SCA) [39]- [41], salp swarm algorithm (SSA) [42]- [44], bat algorithm (BA) [45]- [47], genetic algorithm (GA) [7], [48], [49], flower pollination algorithm (FPA) [50]- [53], cuckoo search (CS) algorithm [54]- [57], differential evolution (DE) [41], [58]- [62], marine predators algorithm (MPA) [63], equilibrium optimizer (EO) [64]- [68], slime mould algorithm (SMA) [69], [70], spotted hyena optimization algorithm [71], [72], emperor penguin optimizer [73], cat swarm optimization algorithm (CSA) [74], [75], harmony search [76]- [81], firefly optimization algorithm (FFA) [82], [83], Chaotic vortex search algorithm [84], crow search algorithm (CSA) [85]- [88], grasshopper optimization algorithm [89]- [92], bacterial foraging algorithm [93]- [95], and dragonfly algorithm [96]- [99].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Improved Binary Grey-Wolf Optimizer With Simulated Annealing for Feature Selection

Abdel‐Basset

Sallam²,

Mohamed

et al. 2021

IEEE Access

View full text Add to dashboard Cite

This paper proposes improvements to the binary grey-wolf optimizer (BGWO) to solve the feature selection (FS) problem associated with high data dimensionality, irrelevant, noisy, and redundant data that will then allow machine learning algorithms to attain better classification/clustering accuracy in less training time. We propose three variants of BGWO in addition to the standard variant, applying different transfer functions to tackle the FS problem. Because BGWO generates continuous values and FS needs discrete values, a number of V-shaped, S-shaped, and U-shaped transfer functions were investigated for incorporation with BGWO to convert their continuous values to binary. After investigation, we note that the performance of BGWO is affected by the selection of the transfer function. Then, in the first variant, we look to reduce the local minima problem by integrating an exploration capability to update the position of the grey wolf randomly within the search space with a certain probability; this variant was abbreviated as IBGWO. Consequently, a novel mutation strategy is proposed to select a number of the worst grey wolves in the population which are updated toward the best solution and randomly within the search space based on a certain probability to determine if the update is either toward the best or randomly. The number of the worst grey wolf selected by this strategy is linearly increased with the iteration. Finally, this strategy is combined with IBGWO to produce the second variant of BGWO that was abbreviated as LIBGWO. In the last variant, simulated annealing (SA) was integrated with LIBGWO to search around the best-so-far solution at the end of each iteration in order to identify better solutions. The performance of the proposed variants was validated on 32 datasets taken from the UCI repository and compared with six wrapper feature selection methods. The experiments show the superiority of the proposed improved variants in producing better classification accuracy than the other selected wrapper feature selection algorithms.INDEX TERMS Grey-wolf optimizer, feature selection, simulated annealing, mutation strategy.

show abstract

“…The use of such strategies means that the search is stuck in a local optima. Finally, we have random search strategies that consist of the application of metaheuristic optimization algorithms [11][12][13][14][15][16][17][18].…”

Section: Introductionmentioning

confidence: 99%

Redundancy Is Not Necessarily Detrimental in Classification Problems

et al. 2021

View full text Add to dashboard Cite

In feature selection, redundancy is one of the major concerns since the removal of redundancy in data is connected with dimensionality reduction. Despite the evidence of such a connection, few works present theoretical studies regarding redundancy. In this work, we analyze the effect of redundant features on the performance of classification models. We can summarize the contribution of this work as follows: (i) develop a theoretical framework to analyze feature construction and selection, (ii) show that certain properly defined features are redundant but make the data linearly separable, and (iii) propose a formal criterion to validate feature construction methods. The results of experiments suggest that a large number of redundant features can reduce the classification error. The results imply that it is not enough to analyze features solely using criteria that measure the amount of information provided by such features.

show abstract

Chaotic vortex search algorithm: metaheuristic algorithm for feature selection

Cited by 98 publications

References 88 publications

Galactic swarm‐improved whale optimization algorithm‐based resource management in Internet of things

Galactic swarm‐improved whale optimization algorithm‐based resource management in Internet of things

An Improved Binary Grey-Wolf Optimizer With Simulated Annealing for Feature Selection

Redundancy Is Not Necessarily Detrimental in Classification Problems

Contact Info

Product

Resources

About