A Multi-Task Decomposition-Based Evolutionary Algorithm for Tackling High-Dimensional Bi-Objective Feature Selection DOI Creative Commons
Hang Xu,

Chaohui Huang,

Jianbing Lin

et al.

Mathematics, Journal Year: 2024, Volume and Issue: 12(8), P. 1178 - 1178

Published: April 14, 2024

Evolutionary algorithms have been widely applied for solving multi-objective optimization problems, while the feature selection in classification can also be treated as a discrete bi-objective problem if attempting to minimize both error and ratio of selected features. However, traditional evolutionary (MOEAs) may drawbacks tackling large-scale selection, due curse dimensionality decision space. Therefore, this paper, we concentrated on designing an multi-task decomposition-based algorithm (abbreviated MTDEA), especially handling high-dimensional classification. To more specific, multiple subpopulations related different tasks are separately initialized then adaptively merged into single integrated population during evolution. Moreover, ideal points these dynamically adjusted every generation, order achieve search preferences directions. In experiments, proposed MTDEA was compared with seven state-of-the-art MOEAs 20 datasets terms three performance indicators, along using comprehensive Wilcoxon Friedman tests. It found that performed best most datasets, significantly better ability promising efficiency.

Language: Английский

Investigating the performance of a surrogate-assisted nutcracker optimization algorithm on multi-objective optimization problems DOI
S. Ida Evangeline,

S. Darwin,

P. Peter Anandkumar

et al.

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 245, P. 123044 - 123044

Published: Dec. 27, 2023

Language: Английский

Citations

4
Enhancing Smart Agriculture Monitoring via Connectivity Management Scheme and Dynamic Clustering Strategy DOI Creative Commons
Fariborz Ahmadi,

Omid Abedi,

Sima Emadi

et al.

Inventions, Journal Year: 2024, Volume and Issue: 9(1), P. 10 - 10

Published: Jan. 5, 2024

The evolution of agriculture towards a modern, intelligent system is crucial for achieving sustainable development and ensuring food security. In this context, leveraging the Internet Things (IoT) stands as pivotal strategy to enhance both crop quantity quality while effectively managing natural resources such water fertilizer. Wireless sensor networks, backbone IoT-based smart agricultural infrastructure, gather ecosystem data transmit them sinks drones. However, challenges persist, notably in network connectivity, energy consumption, lifetime, particularly when facing supernode relay node failures. This paper introduces an innovative approach address these within heterogeneous wireless network-based agriculture. proposed solution comprises novel connectivity management scheme dynamic clustering method facilitated by five distributed algorithms. first second algorithms focus on path collection, establishing connections between each m-supernodes via k-disjoint paths ensure robustness. third fourth provide sustained during failures adjusting transmission powers dynamically sensors based residual energy. fifth algorithm, optimization algorithm implemented dominating set problem strategically position subset nodes migration points mobile supernodes balance network’s depletion. suggested demonstrates superior performance addressing failure tolerance, load balancing, optimal outcomes.

Language: Английский

Citations

1
Automated Plant Disease Detection Systems for the Smart Farming Sector DOI
Priyanga Subbiah, N. Krishnaraj

Advances in environmental engineering and green technologies book series, Journal Year: 2024, Volume and Issue: unknown, P. 296 - 309

Published: Jan. 22, 2024

Global agriculture is affected by plant diseases. Plant diseases have hampered agricultural productivity and development worldwide, reducing food supplies. Systemic conditions can damage leaves. Several were on the The infestation type must be identified to treat it. Farmers' diagnostic error disease propagation are examined in this case study. Machine learning benefit from CV DL methods. This research evaluates dwarf mongoose optimization algorithm with deep for automated leaf detection. APLDD-DMOADL shows farmers photos boost reduce crop losses. method classifies exactly. uses Inception ResNet-v2 extract features stacked LLSTM classify. CSA enhanced subject-level SLSTM hyperparameters. approach was extensively tested using a reference database demonstrate its benefits. Many categories showed that outperformed others.

Language: Английский

Citations

1
Two-stage sparse multi-objective evolutionary algorithm for channel selection optimization in BCIs DOI Creative Commons
Tianyu Liu, Yu Wu,

An Ye

et al.

Frontiers in Human Neuroscience, Journal Year: 2024, Volume and Issue: 18

Published: May 22, 2024

Background Channel selection has become the pivotal issue affecting widespread application of non-invasive brain-computer interface systems in real world. However, constructing suitable multi-objective problem models alongside effective search strategies stands out as a critical factor that impacts performance channel algorithms. This paper presents two-stage sparse evolutionary algorithm (TS-MOEA) to address problems systems. Methods In TS-MOEA, framework, which consists early and late stages, is adopted prevent from stagnating. Furthermore, The two stages concentrate on different models, thereby balancing convergence population diversity TS-MOEA. Inspired by sparsity correlation matrix channels, initialization operator, uses domain-knowledge-based score assignment strategy for decision variables, introduced generate initial population. Moreover, Score -based mutation operator utilized enhance efficiency Results TS-MOEA five other state-of-the-art algorithms been evaluated using 62-channel EEG-based system fatigue detection tasks, results demonstrated effectiveness Conclusion proposed framework can help escape stagnation facilitate balance between convergence. Integrating channels problem-domain knowledge effectively reduce computational complexity while enhancing its optimization efficiency.

Language: Английский

Citations

1
A Multi-Task Decomposition-Based Evolutionary Algorithm for Tackling High-Dimensional Bi-Objective Feature Selection DOI Creative Commons
Hang Xu,

Chaohui Huang,

Jianbing Lin

et al.

Mathematics, Journal Year: 2024, Volume and Issue: 12(8), P. 1178 - 1178

Published: April 14, 2024

Evolutionary algorithms have been widely applied for solving multi-objective optimization problems, while the feature selection in classification can also be treated as a discrete bi-objective problem if attempting to minimize both error and ratio of selected features. However, traditional evolutionary (MOEAs) may drawbacks tackling large-scale selection, due curse dimensionality decision space. Therefore, this paper, we concentrated on designing an multi-task decomposition-based algorithm (abbreviated MTDEA), especially handling high-dimensional classification. To more specific, multiple subpopulations related different tasks are separately initialized then adaptively merged into single integrated population during evolution. Moreover, ideal points these dynamically adjusted every generation, order achieve search preferences directions. In experiments, proposed MTDEA was compared with seven state-of-the-art MOEAs 20 datasets terms three performance indicators, along using comprehensive Wilcoxon Friedman tests. It found that performed best most datasets, significantly better ability promising efficiency.

Language: Английский

Citations

1