| 1. |
- Moravvej, Seyed Vahid, et al.
(författare)
-
An Improved DE Algorithm to Optimise the Learning Process of a BERT-based Plagiarism Detection Model
- 2022
-
Ingår i: 2022 IEEE Congress on Evolutionary Computation (CEC). - : Institute of Electrical and Electronics Engineers (IEEE).
-
Konferensbidrag (refereegranskat)abstract
- Plagiarism detection is a challenging task, aiming to identify similar items in two documents. In this paper, we present a novel approach to automatic plagiarism detection that combines BERT (bidirectional encoder representations from transformers) word embedding, attention mechanism-based long short-term memory (LSTM) networks, and an improved differential evolution (DE) algorithm for weight initialisation. BERT is used to pretrain deep bidirectional representations in all layers, while the pre-trained BERT model can be fine-tuned with only one extra output layer without significant changes in architecture. Deep learning algorithms often use the random weighting method for initialisation, followed by gradient-based optimisation algorithms such as back-propagation for training, making them susceptible to getting trapped in local optima. To address this, population- based metaheuristic algorithms such as DE can be used. We propose an improved DE algorithm with a clustering-based mutation operator, where first a winning cluster of candidate solutions is identified and a new updating strategy is then applied to include new candidate solutions in the current population. The proposed DE algorithm is used in LSTM, attention mechanism, and feed- forward neural networks to yield the initial seeds for subsequent gradient-based optimisation. We compare our proposed model with conventional and population-based approaches on three datasets (SNLI, MSRP and SemEval2014) and demonstrate it to give superior plagiarism detection performance.
|
|
| 2. |
- Al Bataineh, A., et al.
(författare)
-
An efficient hybrid extreme learning machine and evolutionary framework with applications for medical diagnosis
- 2024
-
Ingår i: Expert systems (Print). - : Wiley. - 0266-4720 .- 1468-0394. ; 41:4
-
Tidskriftsartikel (refereegranskat)abstract
- Integrating machine learning techniques into medical diagnostic systems holds great promise for enhancing disease identification and treatment. Among the various options for training such systems, the extreme learning machine (ELM) stands out due to its rapid learning capability and computational efficiency. However, the random selection of input weights and hidden neuron biases in the ELM can lead to suboptimal performance. To address this issue, our study introduces a novel approach called modified Harris hawks optimizer (MHHO) to optimize these parameters in ELM for medical classification tasks. By applying the MHHO-based method to seven medical datasets, our experimental results demonstrate its superiority over seven other evolutionary-based ELM trainer models. The findings strongly suggest that the MHHO approach can serve as a valuable tool for enhancing the performance of ELM in medical diagnosis.
|
|
| 3. |
- Amirsadri, Shima, et al.
(författare)
-
A Levy flight-based grey wolf optimizer combined with back-propagation algorithm for neural network training
- 2017
-
Ingår i: Neural Computing & Applications. - : Springer Nature. - 0941-0643 .- 1433-3058. ; 30:12, s. 3707-3720
-
Tidskriftsartikel (refereegranskat)abstract
- In the present study, a new algorithm is developed for neural network training by combining a gradient-based and a meta-heuristic algorithm. The new algorithm benefits from simultaneous local and global search, eliminating the problem of getting stuck in local optimum. For this purpose, first the global search ability of the grey wolf optimizer (GWO) is improved with the Levy flight, a random walk in which the jump size follows the Levy distribution, which results in a more efficient global search in the search space thanks to the long jumps. Then, this improved algorithm is combined with back propagation (BP) to use the advantages of enhanced global search ability of GWO and local search ability of BP algorithm in training neural network. The performance of the proposed algorithm has been evaluated by comparing it against a number of well-known meta-heuristic algorithms using twelve classification and function-approximation datasets.
|
|
| 4. |
- Bojnordi, Ehsan, et al.
(författare)
-
Improving the Generalisation Ability of Neural Networks Using a Lévy Flight Distribution Algorithm for Classification Problems
- 2023
-
Ingår i: New generation computing. - : Springer Nature. - 0288-3635 .- 1882-7055. ; 41:2, s. 225-242
-
Tidskriftsartikel (refereegranskat)abstract
- While multi-layer perceptrons (MLPs) remain popular for various classification tasks, their application of gradient-based schemes for training leads to some drawbacks including getting trapped in local optima. To tackle this, population-based metaheuristic methods have been successfully employed. Among these, Lévy flight distribution (LFD), which explores the search space through random walks based on a Lévy distribution, has shown good potential to solve complex optimisation problems. LFD uses two main components, the step length of the walk and the movement direction, for random walk generation to explore the search space. In this paper, we propose a novel MLP training algorithm based on the Lévy flight distribution algorithm for neural network-based pattern classification. We encode the network’s parameters (i.e., its weights and bias terms) into a candidate solution for LFD, and employ the classification error as fitness function. The network parameters are then optimised, using LFD, to yield an MLP that is trained to perform well on the classification task at hand. In an extensive set of experiments, we compare our proposed algorithm with a number of other approaches, including both classical algorithms and other metaheuristic approaches, on a number of benchmark classification problems. The obtained results clearly demonstrate the superiority of our LFD training algorithm.
|
|
| 5. |
- Casas-Ordaz, A., et al.
(författare)
-
Enhancing image thresholding segmentation with a novel hybrid battle royale optimization algorithm
- 2024
-
Ingår i: Multimedia tools and applications. - : Springer Nature. - 1380-7501 .- 1573-7721.
-
Tidskriftsartikel (refereegranskat)abstract
- Accurate image segmentation is crucial in digital image processing, enabling efficient image analysis and robust vision systems. However, segmentation is a complex task as images vary in their characteristics, and the computational costs increase with the number of classes involved. To address these challenges, incorporating metaheuristic algorithms to guide the segmentation process presents an exciting opportunity for improvement. This research paper introduces a novel multilevel image segmentation approach that leverages a hybrid battle royale optimization algorithm. By combining opposition-based learning, highly disruptive polynomial mutation, differential evolution mutation, and crossover operators, the proposed method enhances the original battle royale optimization algorithm and effectively solves the segmentation problem. To evaluate the effectiveness of the proposed approach, the minimum cross-entropy criterion is applied to two sets of reference images that undergo multilevel thresholding with up to five thresholds. The results are compared with those obtained using nine other metaheuristic algorithms, employing various image quality metrics such as peak signal noise ratio, structural similarity index method, feature similarity index method, quality index based on local variance, Haar wavelet-based perceptual similarity index, and universal image quality index. The results are analyzed quantitatively, qualitatively, and statistically. The findings demonstrate the potential of the proposed approach in achieving high-quality multilevel thresholding image segmentation. Additionally, the hybrid battle royale optimization algorithm showcases its robustness and efficiency when compared to the other metaheuristic algorithms tested. Notable results are PSNR = 2.13E+01, SSIM = 8.41E-01, FSIM = 8.42E-01, QILV = 8.94E-01, HPSI = 6.52E-01, and UIQI = 9.78E-01.
|
|
| 6. |
- Esmaeili, Leila, et al.
(författare)
-
An efficient method to minimize cross-entropy for selecting multi-level threshold values using an improved human mental search algorithm
- 2021
-
Ingår i: Expert systems with applications. - : Elsevier BV. - 0957-4174 .- 1873-6793. ; 182
-
Tidskriftsartikel (refereegranskat)abstract
- The minimum cross-entropy (MCIT) is introduced as a multi-level image thresholding approach, but it suffers from time complexity, in particular, when the number of thresholds is high. To address this issue, this paper proposes a novel MCIT-based image thresholding based on improved human mental search (HMS) algorithm, a recently proposed population-based metaheuristic algorithm to tackle complex optimisation problems. To further enhance the efficacy, we improve HMS algorithm, IHMSMLIT, with four improvements, including, adaptively selection of the number of mental searches instead of randomly selection, proposing one-step k-means clustering for region clustering, updating based on global and personal experiences, and proposing a random clustering strategy. To assess our proposed algorithm, we conduct an extensive set of experiments with several state-of-the-art and the most recent approaches on a benchmark set of images and in terms of several criteria including objective function, peak signal to noise ratio (PSNR), feature similarity index (FSIM), structural similarity index (SSIM), and stability analysis. The obtained results apparently demonstrate the competitive performance of our proposed algorithm.
|
|
| 7. |
- Helali Moghadam, Mahshid, et al.
(författare)
-
Machine Learning Testing in an ADAS Case Study Using Simulation-Integrated Bio-Inspired Search-Based Testing
- 2022
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This paper presents an extended version of Deeper, a search-based simulation-integrated test solution that generates failure-revealing test scenarios for testing a deep neural network-based lane-keeping system. In the newly proposed version, we utilize a new set of bio-inspired search algorithms, genetic algorithm (GA), (μ+ λ) and (μ,λ) evolution strategies(ES), and particle swarm optimization (PSO), that leverage a quality population seed and domain-specific crossover and mutation operations tailored for the presentation model used for modeling the test scenarios. In order to demonstrate the capabilities of the new test generators within Deeper, we carry out an empirical evaluation and comparison with regard to the results of five participating tools in the cyber-physical systems testing competition at SBST 2021. Our evaluation shows the newly proposed test generators in Deeper not only represent a considerable improvement on the previous version but also prove to be effective and efficient in provoking a considerable number of diverse failure-revealing test scenarios for testing an ML-driven lane-keeping system. They can trigger several failures while promoting test scenario diversity, under a limited test time budget, high target failure severity, and strict speed limit constraints.
|
|
| 8. |
- Helali Moghadam, Mahshid, et al.
(författare)
-
Machine learning testing in an ADAS case study using simulation-integrated bio-inspired search-based testing
- 2024
-
Ingår i: Journal of Software. - : John Wiley and Sons Ltd. - 2047-7473 .- 2047-7481. ; :5
-
Tidskriftsartikel (refereegranskat)abstract
- This paper presents an extended version of Deeper, a search-based simulation-integrated test solution that generates failure-revealing test scenarios for testing a deep neural network-based lane-keeping system. In the newly proposed version, we utilize a new set of bio-inspired search algorithms, genetic algorithm (GA), (Formula presented.) and (Formula presented.) evolution strategies (ES), and particle swarm optimization (PSO), that leverage a quality population seed and domain-specific crossover and mutation operations tailored for the presentation model used for modeling the test scenarios. In order to demonstrate the capabilities of the new test generators within Deeper, we carry out an empirical evaluation and comparison with regard to the results of five participating tools in the cyber-physical systems testing competition at SBST 2021. Our evaluation shows the newly proposed test generators in Deeper not only represent a considerable improvement on the previous version but also prove to be effective and efficient in provoking a considerable number of diverse failure-revealing test scenarios for testing an ML-driven lane-keeping system. They can trigger several failures while promoting test scenario diversity, under a limited test time budget, high target failure severity, and strict speed limit constraints.
|
|
| 9. |
- Hemmati, Majid, et al.
(författare)
-
A New Hybrid Method for Text Feature Selection Through Combination of Relative Discrimination Criterion and Ant Colony Optimization
- 2022
-
Ingår i: Proceedings of 7th International Conference on Harmony Search, Soft Computing and Applications. - Singapore : Springer Nature.
-
Konferensbidrag (refereegranskat)abstract
- Text categorization plays a significant role in many information management tasks. Due to the increasing volume of documents on the Internet, automated text categorization has been more considered for classifying documents in pre-defined categories. A major problem of text categorization is the high dimensionality of feature space. Most of the features are irrelevant and redundant impacting the classifier performance. Hence, feature selection is used to reduce the high dimensionality of feature space and increase classification efficiency. In this paper, we proposed a hybrid two-stage method for text feature selection based on Relative Discrimination Criterion (RDC) and Ant Colony Optimization (ACO). To this end, we applied RDC method, at first, in order to rank features based on their values. Features, then, which their values are lower than a threshold are removed from the feature set. In the second stage, as a wrapper method, an ACO-based feature selection method is applied, to select redundant or irrelevant features that have not been removed in the first stage. Finally, to assess the proposed methods, we have conducted several experiments on different datasets to indicate the superiority of our proposed algorithm. We aim to propose a hybrid approach which is computationally more efficient in much the same way as it is more accurate compared to the other embedded or wrapper methods. The obtained results endorse that the proposed method is of remarkable performance in text feature selection.
|
|
| 10. |
- lehi, Arash Yunessnia, et al.
(författare)
-
Pre-treatment of textile wastewaters containing Chrysophenine using hybrid membranes
- 2017
-
Ingår i: Membrane Water Treatment. - : Techno-Press. - 2005-8624. ; 8:1, s. 89-112
-
Tidskriftsartikel (refereegranskat)abstract
- Dyeing wastewaters are the most problematic wastewater in textile industries and also, growing amounts of waste fibers in carpet industries have concerned environmental specialists. Among different treatment methods, membrane filtration processes as energy-efficient and compatible way, were utilized for several individual problems. In this research, novel hybrid membranes were prepared by waste fibers of mechanical carpets as useful resource of membrane matrix and industrial graphite powder as filler to eliminate Chrysophenine GX from dyeing wastewater. These membranes were expected to be utilized for first stage of hybrid membrane filtration process including (adsorption-ultrafiltration) and nanofiltration in Kashan Textile Company. For scaling of membrane filtration process, fouling mechanism of these membranes were recognized and explained by the use of genetic algorithm, as well. The graphite increased rejection and diminished permeate flux at low concentration but in high concentration, the performance was significantly worsened. Among all hybrid membranes, 18% wt. waste fibers-1% wt. graphite membrane had the best performance and minimum fouling. The maximum pore size of this optimum membrane was ranged from 16.10 to 18.72 nm
|
|
