NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION
The Artificial Neural Networks Training (ANNT) process is an optimization problem of the weight set which has inspired researchers for a long time. By optimizing the training of the neural networks using optimal weight set, better results can be obtained by the neural networks. Traditional neural ne...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
UUM Press
2017-11-01
|
| Series: | Journal of ICT |
| Subjects: | |
| Online Access: | https://e-journal.uum.edu.my/index.php/jict/article/view/8234 |
| _version_ | 1818490201032359936 |
|---|---|
| author | Zakaria Noor Aldeen Mahmood Al Nuaimi Rosni Abdullah |
| author_facet | Zakaria Noor Aldeen Mahmood Al Nuaimi Rosni Abdullah |
| author_sort | Zakaria Noor Aldeen Mahmood Al Nuaimi |
| collection | DOAJ |
| description | The Artificial Neural Networks Training (ANNT) process is an optimization problem of the weight set which has inspired researchers for a long time. By optimizing the training of the neural networks using optimal weight set, better results can be obtained by the neural networks. Traditional neural networks algorithms such as Back Propagation (BP) were used for ANNT, but they have some drawbacks such as computational complexity and getting trapped in the local minima. Therefore, evolutionary algorithms like the Swarm Intelligence (SI) algorithms have been employed in ANNT to overcome such issues. Artificial Bees Colony (ABC) optimization algorithm is one of the competitive algorithms in the SI algorithms group. However, hybrid algorithms are also a fundamental concern in the optimization field, which aim to cumulate the advantages of different algorithms into one algorithm. In this work, we aimed to highlight the performance of the Hybrid Particle-move Artificial Bee Colony (HPABC) algorithm by applying it on the ANNT application. The performance of the HPABC algorithm was investigated on four benchmark pattern-classification datasets and the results were compared with other algorithms. The results obtained illustrate that HPABC algorithm can efficiently be used for ANNT. HPABC outperformed the original ABC and PSO as well as other state-of-art and hybrid algorithms in terms of time, function evaluation number and recognition accuracy.
|
| first_indexed | 2024-12-10T17:14:08Z |
| format | Article |
| id | doaj.art-80c31aeda360479380dfe7caa95a88cb |
| institution | Directory Open Access Journal |
| issn | 1675-414X 2180-3862 |
| language | English |
| last_indexed | 2024-12-10T17:14:08Z |
| publishDate | 2017-11-01 |
| publisher | UUM Press |
| record_format | Article |
| series | Journal of ICT |
| spelling | doaj.art-80c31aeda360479380dfe7caa95a88cb2022-12-22T01:40:11ZengUUM PressJournal of ICT1675-414X2180-38622017-11-01162NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATIONZakaria Noor Aldeen Mahmood Al Nuaimi0Rosni Abdullah1School of Computer Sciences Universiti Sains Malaysia, MalaysiaSchool of Computer Sciences Universiti Sains Malaysia, MalaysiaThe Artificial Neural Networks Training (ANNT) process is an optimization problem of the weight set which has inspired researchers for a long time. By optimizing the training of the neural networks using optimal weight set, better results can be obtained by the neural networks. Traditional neural networks algorithms such as Back Propagation (BP) were used for ANNT, but they have some drawbacks such as computational complexity and getting trapped in the local minima. Therefore, evolutionary algorithms like the Swarm Intelligence (SI) algorithms have been employed in ANNT to overcome such issues. Artificial Bees Colony (ABC) optimization algorithm is one of the competitive algorithms in the SI algorithms group. However, hybrid algorithms are also a fundamental concern in the optimization field, which aim to cumulate the advantages of different algorithms into one algorithm. In this work, we aimed to highlight the performance of the Hybrid Particle-move Artificial Bee Colony (HPABC) algorithm by applying it on the ANNT application. The performance of the HPABC algorithm was investigated on four benchmark pattern-classification datasets and the results were compared with other algorithms. The results obtained illustrate that HPABC algorithm can efficiently be used for ANNT. HPABC outperformed the original ABC and PSO as well as other state-of-art and hybrid algorithms in terms of time, function evaluation number and recognition accuracy. https://e-journal.uum.edu.my/index.php/jict/article/view/8234Swarm IntelligenceArtificial Neural NetworksArtificial Bee Colony AlgorithmParticle Swarm OptimizationPattern-Classification |
| spellingShingle | Zakaria Noor Aldeen Mahmood Al Nuaimi Rosni Abdullah NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION Journal of ICT Swarm Intelligence Artificial Neural Networks Artificial Bee Colony Algorithm Particle Swarm Optimization Pattern-Classification |
| title | NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION |
| title_full | NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION |
| title_fullStr | NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION |
| title_full_unstemmed | NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION |
| title_short | NEURAL NETWORK TRAINING USING HYBRID PARTICLEMOVE ARTIFICIAL BEE COLONY ALGORITHM FOR PATTERN CLASSIFICATION |
| title_sort | neural network training using hybrid particlemove artificial bee colony algorithm for pattern classification |
| topic | Swarm Intelligence Artificial Neural Networks Artificial Bee Colony Algorithm Particle Swarm Optimization Pattern-Classification |
| url | https://e-journal.uum.edu.my/index.php/jict/article/view/8234 |
| work_keys_str_mv | AT zakarianooraldeenmahmoodalnuaimi neuralnetworktrainingusinghybridparticlemoveartificialbeecolonyalgorithmforpatternclassification AT rosniabdullah neuralnetworktrainingusinghybridparticlemoveartificialbeecolonyalgorithmforpatternclassification |