Backpropagation Neural Network optimization and software defect estimation modelling using a hybrid Salp Swarm optimizer-based Simulated Annealing Algorithm

Kassaymeh, Sofian; Al-Laham, Mohamad; Al-Betar, Mohammed Azmi; Alweshah, Mohammed; Abdullah, Salwani; Makhadmeh, Sharif Naser

doi:10.1016/j.knosys.2022.108511

Cited by 39 publications

(15 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Another example is the hybridization of Whale Optimization Algorithm with SA to improve the WOA exploitation for feature selection [72]. Also, the hybridization of the Salp Swarm Algorithm (SSA) with SA Algorithm to adjust the balance between exploration and exploitation of SSA algorithm [73]. Finally, Monarch Butterfly Optimization (MBO) with SA strategy to improve the convergence speed of MBO algorithm.…”

Section: Related Workmentioning

confidence: 99%

Improved Reptile Search Optimization Algorithm Using Chaotic Map and Simulated Annealing for Feature Selection in Medical Field

et al. 2022

Self Cite

View full text Add to dashboard Cite

The increased volume of medical datasets has produced high dimensional features, negatively affecting machine learning (ML) classifiers. In ML, the feature selection process is fundamental for selecting the most relevant features and reducing redundant and irrelevant ones. The optimization algorithms demonstrate its capability to solve feature selection problems. Reptile Search Algorithm (RSA) is a new nature-inspired optimization algorithm that stimulates Crocodiles' encircling and hunting behavior. The unique search of the RSA algorithm obtains promising results compared to other optimization algorithms. However, when applied to high-dimensional feature selection problems, RSA suffers from population diversity and local optima limitations. An improved metaheuristic optimizer, namely the Improved Reptile Search Algorithm (IRSA), is proposed to overcome these limitations and adapt the RSA to solve the feature selection problem. Two main improvements adding value to the standard RSA; the first improvement is to apply the chaos theory at the initialization phase of RSA to enhance its exploration capabilities in the search space. The second improvement is to combine the Simulated Annealing (SA) algorithm with the exploitation search to avoid the local optima problem. The IRSA performance was evaluated over 20 medical benchmark datasets from the UCI machine learning repository. Also, IRSA is compared with the standard RSA and stateof-the-art optimization algorithms, including Particle Swarm Optimization (PSO), Genetic Algorithm (GA), Grasshopper Optimization algorithm (GOA) and Slime Mould Optimization (SMO). The evaluation metrics include the number of selected features, classification accuracy, fitness value, Wilcoxon statistical test (pvalue), and convergence curve. Based on the results obtained, IRSA confirmed its superiority over the original RSA algorithm and other optimized algorithms on the majority of the medical datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Improved Reptile Search Optimization Algorithm Using Chaotic Map and Simulated Annealing for Feature Selection in Medical Field

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…The input training set for softmax regression with v number of data vector {(x1, y1), (x2, y2), …(xv, yv) [21], [22]. In the softmax regression-based classifier, the probability P (Y =j | X) of X belonging to each class from a set of k classes is given as shown in the formula (11). This softmax regression cost function has no closed form way to minimize the cost value, so the iterative algorithm, gradient descent, is used, which is shown in formula (13):…”

Section: Softmax Regressionmentioning

confidence: 99%

“…RBM layers will be used for unsupervised learning. To perform better training, the setting in full node and learning rate is essential [11], [12]. After that, the model will be used for supervised learning using softmax regression in the output layer.…”

Section: Introductionmentioning

confidence: 99%

Restricted boltzmann machine and softmax regression for acute respiratory infections disease identification

2022

joscex

View full text Add to dashboard Cite

Restricted boltzmann machines (RBM) have attracted much attention lately after being proposed as building blocks of deep learning blocks. RBM is an algorithm that belongs to the artificial neural network (ANN) algorithm. Deep learning models can be used in the health field to identify diseases using medical data records. Acute Respiratory Infection (ARI) is a disease that infects the respiratory tract. A patient infected by ARI diseases is high. To identify ARI can use the symptoms that the patient had experienced. Based on this background, this study aims to help identify ARI disease using its symptoms. The method used for identification is the deep learning model, which was built using the RBM and softmax regression. Three steps were used in this research, which are training, testing, and implementation. The trained deep learning model will be implemented to identify ARI disease. This research will use ARI data from Puskemas Warungasem, Indonesia. From the research result, the deep learning model can get an accuracy of 96%. The deep learning configuration used in this research has 4 RBM layers, 1 Softmax layer as the output layer, and a learning rate value of 0.01 and 1000 iterations. This research can be used as a reference so that the next researcher can add other algorithms to Deep learning to improve accuracy.

show abstract

“…Untuk mendapatkan error ini, tahap forward propagation harus dilakukan terlebih dahulu. Saat propagasi berlangsung, neuron diaktifkan menggunakan fungsi aktivasi yang dapat dibedakan seperti sigmoid [14].…”

Section: Time Series Prediction Dengan Backpropagationunclassified

Smart Farming Untuk Pengaturan Suhu Ruangan Pada Budidaya Jamur Tiram Berbasis Backpropagation

Sugiartawan

Desnanjaya

2022

Indonesian J. Electron. Instrum. Syst.

View full text Add to dashboard Cite

The problem with mushroom cultivation is the difficulty of regulating the room temperature of mushrooms, especially oyster mushrooms. The optimal production of oyster mushrooms is at temperatures between 25 C - 27 C. To regulate or manipulate humidity and room temperature to water the kumbung or mushroom room. The watering process is carried out several times to stabilize the room temperature during the day.To overcome the watering that is done manually, Automatic Temperature Control and Monitoring of Oyster Mushrooms Based on GSM Sim800l Arduino Uno is made. This tool uses a DHT11 sensor, relay, 16x2 LCD, GSM Sim 800L, and Stepdown. The test was carried out in a mushroom kumbung measuring 10.7m long, 5.9m wide, and 3.5m high. Watering time is done by observing the data at room temperature. The data is then studied using a backpropagation. This method aims to identify the pattern of watering time so that the optimal watering time is produced. The test results show that the tool can monitor the temperature and humidity of the kumbung mushroom with the following values: temperature 27°C - 33°C and humidity 70% - 90%. The introduction of mushroom watering patterns with BPNN showed an error rate of 40%.

show abstract

Backpropagation Neural Network optimization and software defect estimation modelling using a hybrid Salp Swarm optimizer-based Simulated Annealing Algorithm

Cited by 39 publications

References 61 publications

Improved Reptile Search Optimization Algorithm Using Chaotic Map and Simulated Annealing for Feature Selection in Medical Field

Improved Reptile Search Optimization Algorithm Using Chaotic Map and Simulated Annealing for Feature Selection in Medical Field

Restricted boltzmann machine and softmax regression for acute respiratory infections disease identification

Smart Farming Untuk Pengaturan Suhu Ruangan Pada Budidaya Jamur Tiram Berbasis Backpropagation

Contact Info

Product

Resources

About