Memetic differential evolution methods for clustering problems

Mansueto, Pierluigi; Schoen, Fabio

doi:10.1016/j.patcog.2021.107849

Cited by 21 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the work of [58,8,34,35], DC (Difference of Convex functions) programming is used to define efficient heuristic algorithms for clustering large datasets. The algorithm k-means has also been used as a local search subroutine in different algorithms, as in the population-based metaheuristic developed in [27] and in the differential evolution scheme proposed in [44].…”

Section: Literature Reviewmentioning

confidence: 99%

SOS-SDP: an Exact Solver for Minimum Sum-of-Squares Clustering

Piccialli,

Sudoso,

Wiegele

2021

Preprint

View full text Add to dashboard Cite

The minimum sum-of-squares clustering problem (MSSC) consists in partitioning n observations into k clusters in order to minimize the sum of squared distances from the points to the centroid of their cluster. In this paper, we propose an exact algorithm for the MSSC problem based on the branch-andbound technique. The lower bound is computed by using a cutting-plane procedure where valid inequalities are iteratively added to the Peng-Wei SDP relaxation. The upper bound is computed with the constrained version of k-means where the initial centroids are extracted from the solution of the SDP relaxation. In the branch-and-bound procedure, we incorporate instance-level must-link and cannot-link constraints to express knowledge about which instances should or should not be grouped together. We manage to reduce the size of the problem at each level preserving the structure of the SDP problem itself. The obtained results show that the approach allows to successfully solve for the first time real-world instances up to 4000 data points.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

SOS-SDP: an Exact Solver for Minimum Sum-of-Squares Clustering

Piccialli,

Sudoso,

Wiegele

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…To overcome the low search ability and premature convergence of quantum inspired differential evolution (QDE), an enhanced MSIQDE algorithm was proposed in [65]. For solving Minimum Sum-of-Squares Clustering problem, Memetic Differential Evolution (MDE) was employed in [66]. The authors offered four algorithm variants that differ on the matching approach.…”

Section: Implementation Of Optimization Algorithm: Differential Evolutionmentioning

confidence: 99%

Optimal Planning of Integrated Nuclear-Renewable Energy System for Marine Ships Using Artificial Intelligence Algorithm

Adham

Abdussami

2021

Energies

View full text Add to dashboard Cite

Ocean-going ships are one of the primary sources of Greenhouse Gas (GHG) emissions. Several actions are being taken to reduce the GHG emissions from maritime vessels, and integration of Renewable Energy Sources (RESs) is one of them. Ocean-going marine ships need a large amount of reliable energy to support the propulsive load. Intermittency is one of the drawbacks of RESs, and penetration of RESs in maritime vessels is limited by the cargo carrying capacity and usable area of that ship. Other types of reliable energy sources need to be incorporated in ships to overcome these shortcomings of RESs. Some researchers proposed to integrate fossil fuel-based generators like diesel generators and renewable energy in marine vessels to reduce GHG emissions. As the penetration of RESs in marine ships is limited, fossil fuel-based generators provide most of the energy. Therefore, renewable and fossil fuel-based hybrid energy systems in maritime vessels can not reduce GHG emissions to the desired level. Fossil fuel-based generators need to be replaced by emissions-free energy sources to make marine ships free from emissions. Nuclear energy is emissions-free energy, and small-scale nuclear reactors like Microreactors (MRs) are competent to replace fossil fuel-based generators. In this paper, the technical, environmental, and economic competitiveness of Nuclear-Renewable Hybrid Energy Systems (N-R HES) in marine ships are assessed. The lifecycle cost of MR, reliability of the proposed system, and limitations of integrating renewable energy in maritime vessels are considered in this study. The proposed N-R HES is compared with three different energy systems, namely ‘Standalone Fossil Fuel-based Energy Systems’, ‘Renewable and Fossil Fuel-based Hybrid Energy Systems’, and ‘Standalone Nuclear Energy System’. The cost modeling of each energy system is carried out in MATLAB simulator. Each energy system is optimized by using the Differential Evolution Algorithm (DEA), an artificial intelligence algorithm, to find out the optimal configuration of the system components in terms of Net Present Cost (NPC). The results determine that N-R HES has the lowest NPC compared to the other three energy systems. The performance of the DE algorithm is compared with another widely accepted artificial intelligence optimization technique called ‘Particle Swarm Optimization (PSO)’ to validate the findings of the DE algorithm. The impact of control parameters in the DE algorithm is assessed by employing the Adaptive Differential Evolution (ADE) algorithm. A sensitivity analysis is carried out to assess the impact of different system parameters on this study’s findings.

show abstract

“…Jain and Dharavath have proposed memetic salp swarm algorithm for feature extraction in crop disease detection system 30 . Mansueto and Schoen have presented a new implementation of the memetic differential evolution (MDE) algorithm based on the development of k‐means for the Euclidean minimum sum‐of‐squares clustering clustering problem 31 . Similarly, Mustafa et al have used MDE to solve text clustering problems 32 …”

Section: Introductionmentioning

confidence: 99%

“…30 Mansueto and Schoen have presented a new implementation of the memetic differential evolution (MDE) algorithm based on the development of k-means for the Euclidean minimum sum-of-squares clustering clustering problem. 31 Similarly, Mustafa et al have used MDE to solve text clustering problems. 32 Like memetic algorithms, another approach inspired by GA is the elite evolution strategy (ES).…”

Section: Introductionmentioning

confidence: 99%

Marine predator algorithm with elite strategies for engineering design problems

Aydemı̇r

Onay

2023

Concurrency and Computation

View full text Add to dashboard Cite

Marine predator algorithm (MPA) is a powerful metaheuristic optimization algorithm that shows effective convergence ability on complex benchmark functions. The combination of Brownian and Levy flight distributions directly affects the convergence strategy of MPA. Although MPA has good convergence performance, it is open to improvement as it falls to a local optimum and cannot comprehensively scan the search area during the exploration phase. In this study, MPA has been improved by integrating elite natural evolution and elite random mutation strategies. In addition, these two strategies are combined with Gaussian mutation. The proposed method in this study which is named as elite evolution strategy MPA (EEMPA) has achieved comprehensive scanning of the solution space and considerably reduced the risk of falling into the local optimum trap, with elite strategies. The effect of EEMPA has been tested with the CEC2017 and CEC2019 benchmark functions. EEMPA has been compared with some metaheuristic algorithms frequently used in the literature and gives promising results among the considered optimization methods. Furthermore, EEMPA has been examined for seven well-known real world engineering problems. When the results are compared with both classical MPA and enhanced MPA methods, EEMPA converges to better than the other methods.

show abstract

Memetic differential evolution methods for clustering problems

Cited by 21 publications

References 25 publications

SOS-SDP: an Exact Solver for Minimum Sum-of-Squares Clustering

SOS-SDP: an Exact Solver for Minimum Sum-of-Squares Clustering

Optimal Planning of Integrated Nuclear-Renewable Energy System for Marine Ships Using Artificial Intelligence Algorithm

Marine predator algorithm with elite strategies for engineering design problems

Contact Info

Product

Resources

About