Description Authors Metaheuristic algorithm

proposes the hybridization of a Differential Evolution DE algorithm with an ANN-based regression as a way to apply the local search heuristic, as a predictor of the best solution from a training set of trial vectors produced by an ensemble of DE strategies. Fister, Iztok, et al., 2016 Differential Evolution

(DE)

Proposes a hybrid Chemical Reaction Optimizations (CRO) algorithm-based application using the DE mutation strategies in the training of the Higher Order Neural Networks (HONNs), mainly the Pi-Sigma Network (OSN). Sibarama, 2017 Presented a hybrid novel multi-Objective Neural Network with Differential Evolution (MONNDE) model, which proposed to investigate evolving Neural Network capable of producing pareto front for dynamic multi-objective optimisation problem. Mason, K., Duggan, J., & Howley, E. (2018) proposed a particle swarm optimization-based approach to train the NN (NN-PSO). The PSO is employed to find a weight vector for NN. The proposed (NN-PSO) classifier is capable to tackle the problem of predicting structural failure of multistoried reinforced concrete buildings via detecting the failure possibility of the multistoried RC building structure in the future. Chatterjee, Sankhadeep, et al., 2017 Particle Swarm Optimization

(PSO)

Developed two intelligent systems namely artificial neural network (ANN) and particle swarm optimization (PSO)–ANN models to predict factor of safety (FOS) of homogeneous slopes. Gordan, Behrouz, et al., 2016 presented a novel, Particle Swarm Optimization (PSO)-trained Quantile Regression Neural Network namely PSOQRNN, to forecast volatility from financial time series. as observed that the proposed PSOQRNN yielded statistically significant results compared to other popular volatility forecasting models Pradeepkumar, D., & Ravi, V. (2017) Proposed a hybrid ANN-PSO forecasting model for forecasting the groundwater level in Udupi district. The results reveal that PSO-ANN based hybrid model gives a better prediction accuracy, than ANN alone. Balavalikar, Supreetha, et al., 2018 introduced a method named as Bacterial foraging optimization based Radial Basis Function Neural Network (BRBFNN) for identification and classification of plant leaf diseases automatically. using Bacterial foraging optimization (BFO) For assigning optimal weight to Radial Basis Function Neural Network (RBFNN). The use of this model gave impressive results in the accuracy of classification and identification of diseases. Shi, J. (2017) Bacterial foraging optimization

(BFO)

presented an innovative automatic ( BFO-ANN) model to diagnose the diabetes disease. To discard the irrelevant features BFO was used, and then ANN was applied to the selected features to find out the classification accuracy. Sharma, R. P. M., & SSCET, B. (2016) proposed a hybrid approach integrating the Bacterial Foraging Optimization Algorithm (BFOA) in a Radial Basis Function Neural Network (RBFNN), applied to image classification, in order to improve the classification rate and the objective function value. It very robust and efficient for several image classification. Amghar, Y. T., & Fizazi, H. (2017) prediction method based on artificial Fish Swarm Algorithm (FSA) and RBF Neural Network (RBFNN) is proposed to predict the stock price trend. The simulation shows that, the proposed method is better than BPNN and RBFNN in prediction accuracy for stock price trend. It provides an effective and feasible method for stock price prediction. Yanming, W., Xusheng, G., & Lei, L. (2017) Artificial Fish Swarm Algorithm (AFSA)

proposed a new classifier based on artificial neural networks (ANN) trained using artificial fish swarm optimization (AFSO) algorithm for improving the sensitivity and specificity of DCE-MR breast image interpretation. The AFSO algorithm was used for searching the best combination of synaptic weights for the neural network Sathya, D. J., & Geetha, K. (2017) Applied an Artificial Neural Network (ANN) and the Bees Algorithm (BA) to model and optimize the continuous synthesis of high purity carbon nanotubes ( CNTs) production. The computational results have relatively good agreement with the experimental results. Ahangarpour, Ameneh, et al., 2017 Bees Algorithm

(BA)

Proposed a new Artificial Neural Networks with bees algorithm (ANN-BA) for optimization of a low-density extraction solvent-based ST-DLLME of Cu2+ ions with di2-ethylhexylphosphoric acid. And central composite design (CCD) was also used as a comparative method for optimization of extraction process. Farajvand, Mohammad, et al., 2017 Presented a hybrid multi-layer perceptron neural network and Bees Algorithm MLPNN-BA method. For torque control in wind turbines for Generating maximum possible power by wind turbines at low wind speeds. The results of simulations indicate the appropriate performance of the proposed method. Assareh, E., & Biglari, M. (2016) Proposed a new optimal adaptive controller using Bees Algorithm(BA), Artificial Neural Network (ANN) for active front steering control of vehicles by adjusting the controller (PID) coefficients for any arbitrary road conditions. Aalizadeh, B., & Asnafi, A. (2016) Developed an innovative intelligent optimization model based on nonlinear support vector machine and glowworm swarm optimization to forecast daily global solar radiation and to discover association rules between solar radiation and several meteorological factors. Results reveal that the proposed model delivers the best forecasting performances comparing with other competitors. Jiang, H., & Dong, Y. (2016) Glowworm Swarm Optimization (GSO)

Proposed a data fusion algorithm based on glowworm swarm algorithm optimized back propagation (GSO-BP) neural network (NN) . To improve the data transmission efficiency of wireless sensor networks (WSNs), and reduce the data traffic and the energy consumption of the sensing nodes in the network. Results showed that the proposed algorithm can efficiently reduce the transmission traffic of networks and the energy consumption of sensing nodes. Wu, W., Xu, B., & Cao, M. (2016) Established diagnosis model Glowworm swarm Optimization with support vector machine GSO-SVM to diagnosis hydropower unit vibration fault. The simulation results show that the established model has better convergence speed , global optimization ability and higher calculation precision. Zheng, X., Gui, Z., & Wang, Y. (2017) glowworm optimization technique and support vector machine work combined and improved the performance of cyber-attack detection. By applying the reduction feature process to improve the capacity classification and detection of attack. Developed a new model ,Improved Shuffled Frog Leaping Algorithm and BP Neural Network

ISFLA-BP model, for the early fault diagnosis of rolling bearings. ISFLA is employed to optimize the weights and threshold values of BP neural network. Results indicate that the developed model is able to effectively improve the efficiency of network training and the accuracy of early fault pattern recognition in bearing fault diagnosis tasks. Zhao, Z., Xu, Q., & Jia, M. (2016) Shuffled Frog Leaping Algorithm

(SFLA)

Presented a novel predictor model using Pi-Sigma (higher order neural network) with Improved Shuffled Frog Leaping Algorithm (ISFLA) for accurate and unbiased prediction of future currency exchange rate . Practical analysis of results suggests that the Pi-Sigma network learned with ISFL is a promising predictor model for currency exchange rate prediction Dash, R. (2018) Proposed a new forecasting model MSFLA-LSSVM , least squares support vector machine (LSSVM) modified shuffled frog leaping (MSFLA) algorithm, to forecast the CO2 emissions in China accurately according to several factors that increase the emission problem. Precise results led to the adoption of this model for the next seven years. Dai, S., Niu, D., & Han, Y. (2018) proposed a novel iris recognition system using FFBNN-ACFO , Feed Forward Back Propagation Neural Network and Adaptive Central Force Optimization, After removing the noise, features extracted from the image are used to train the neural network whose parameters are tuned by the ACFO to obtain high recognition accuracy. Shaikh, N. F., & Doye, D. D. (2016) Central Force Optimization

(CFO)

developed an optimized (FPSOGSA) model, NN with fuzzy hybrid of Particle Swarm Optimization and Gravitational Search Algorithm , that could be deployed in monitoring and tracking devices used for locating users based on the Wi-Fi signal strength they receive in their personal devices. Rohra, Jayant G., et al., 2017 Gravitational Search Algorithm

(GSA)

Presented An innovative algorithm which avoid weaknesses in BPNN and GSA algorithms (i.e., the process of selecting optimal parameters and avoiding local optima trap) for dealing with Multinomial Classification. Validation of this method is done by applying it on benchmark datasets and comparing the results. Priyadarshini, R., Panda, M. R., & Dash, N. (2018) introduced an iris classification system by training feed-forward Neural Network using (GSA) ( FFNNGSA) and Particle Swarm Optimization (PSO) FFNNPSO). Some techniques are applied to the image and extract the features from them and then the algorithms GSA and PSO are used to training the FFNN to get the optimal weights and biases. The results showed that GSA is better than PSO in FFNN training. Rizk, M. R., Farag, H. H., & Said, L. A. (2016) Proposed a n optimal classifier which combined MGSO and Fuzzy Min-Max Neural Network (GFMMNN) to classify the medical data with high accuracy value. Due to the high-dimensional input data, an OLPP was used to reduce the features and use only useful ones. Results shows high classification accuracy. Rafi, D. M., & Bharathi, C. R. (2016) Group Search Optimizer

(GSO)

Proposed a backcalculation algorithm, named LCA-ANN, based on the combined use of metaheuristic optimization method called League Championship Algorithm (LCA) and finite element method based Artificial Neural Network (ANN), for backcalculating layer properties of conventional flexible pavements .results indicate that LCA-ANN can be used as a robust backcalculation method. Tezcan, B. M., & Pekcan, O. (2017) League Championship Algorithm

(LCA)

investigated a new classification approach , (LCA-FCRN), using League Championship Algorithm Optimized ensembled Fully Complex valued Relaxation Network for detection of breast abnormalities in digital mammograms. Proposed method can form an effective CAD system, and achieve good classification accuracy. Saraswathi, D., & Srinivasan, E. (2017) Proposed a new combined method , Artificial Neural Networks (ANN) and Eagle Strategy (ES) , for efficient shape optimization method for centrifugal pump. Furthermore the ES and PSO algorithm was compared and results shows that ES is efficient than PSO algorithm in this application and this methodology is more efficient than other surrogate methods. Derakhshan, S., & Bashiri, M. (2018) Eagle Strategy

Proposed a novel method using Artificial Neural Network (ANN) optimized by Fireworks Algorithm (FWA). Aimed to ameliorate the accuracy of agile software effort prediction. results indicated that the FWA is an effective algorithm to improve the accuracy of the ANN in comparison with the other algorithms ,and ANN also outperformed the other types of ANN. Khuat, T. T., & Le, M. H. (2016) Fireworks algorithm

(FWA)

proposed a novel application, where Fireworks Algorithm(FWA) was applied in training of Multi-Layer Perceptron(MLP) for classification task in medical data mining. Comparing (FWA) with PSO-W an LM algorithms , FWA performs better in classification task and also maintains a good trade-off between sensitivity and specificity. Dutta, R. K., Karmakar, N. K., & Si, T. (2016) Proposed a new method based on fireworks algorithm (FWA) for the training ANNs, aimed for classification purpose. FWA was better than others in neural network training because of its advantages in balancing exploration and exploitation that make it an effective optimizer. Bolaji, A. L. A., Ahmad, A. A., & Shola, P. B. (2016) Proposed a novel hybrid optimization algorithm based on Bayesian classifier and Hunting Search (HuS) algorithm, Which extracts the best features from the data, for Epileptic real-time seizure prediction. MLP has been trained with the optimal features to detect the seizure attacks in the online mode. Behnam, M., & Pourghassem, H. (2016) proposed a new predictor model based on Flower Pollination Algorithm (FPA) and NN for the forecasting of petroleum consumption by the Organization of the Petroleum Exporting Countries (OPEC). The method proposed performs better than the GA, ABC, APSO and BP in terms of convergence speed, forecasting accuracy. Chiroma, Haruna, et al., 2016 Flower Pollination algorithm

(FPA)

proposed a new technique for feature selection and classification of breast cancer, FPA-SVM, based on Flower Pollination algorithm (FPA) and Support Vector machine (SVM) using microarray data. Dankolo, Muhammad Nasiru, et al., 2018 Investigated the effectiveness of neural network training by Krill Herd Algorithm (KHA).

The efficiency of this algorithm was tested in the training with the comparison of the evaluation criteria. And the results showed remarkable efficacy in terms of training time and evaluation measures. Kowalski, P. A., & ?ukasik, S. (2016) Krill Herd Algorithm

(KHA)

Assessed the performance of the krill herd algorithm (KHA) in the training of the Feed-Forward Neural Network (FFNN) for finding the best weights. By comparing it with the genetic algorithm (GA) ,Back Propagation Neural Network( BPNN) , and linear regression model, it has higher efficiency in optimizing and optimizing the FFNN Asteris, Panagiotis G., et al., 2018 Presented a new predictor model using Neural Network(NN) and Black Hole Algorithm (BHA) To predict the workload of the cloud computing system. The model achieved a greater improvement in the average square error than in the BP. Kumar, J., & Singh, A. K. (2016) Black Hole Algorithm (BHA)

Proposed an improved technique to solve engineering optimization problems using adaptive neuro-fuzzy inference system(ANFIS) that trained with Accelerated Mine Blast Algorithm (IMBA), Which replaced the MBA to increase the speed of convergence and gain more accurate results. Salleh, M., Najib, M., & Hussain, K. (2016) Mine Blast Algorithm (MBA)

Presented a speed controller of separately excited DC motor to train the NN parameters . NN was used to predict the field current that realizes the field weakening to drive the motor over rated speed. Results showed that the proposed method gives a good performance and is feasible to be applied instead of others conventional combined control methods Hameed, W. I., Kadhim, A. S., & Al-Thuwaynee, A. A. K. (2016) Social Spider Optimization

(SSO)

Presented a new method for predicting the production of biochar by combining the Adaptive Neuro-Fuzzy Inference System (ANFIS) and Social-Spider Optimization algorithm (SSO). This method was compared with classical ANFIS, ABC, PSO, and LS-SVM, the results indicated the superiority of this method on other methods. E wees, A. A., El Aziz, M. A., & Elhoseny, M. (2017) Proposed a hybrid technique , GWO-MLP, To detect the types of skin cancer through the images, was used some of the techniques of image processing on the images and then apply this technique to be segmented. The GWO performed very well in improving MLP performance. Parsian, A., Ramezani, M., & Ghadimi, N. (2017). Gray Wolf Optimizer (GWO)

Introduced a hybrid optimized classification method (GWO–ANN) to distinguish brain tumors by magnetic resonance images (MRI) as natural images or otherwise. This method provided a much better rating and performance rate than traditional neural network use. Ahmed, Heba M., et al., 2018 Proposed GWO-ANN technique to deal with system identification problems. GWO was used With the aim of training ANN parameters. This technique is capable of dealing with identification problems as the results show. Ramezani, M., & Motlagh, M. M. (2016) Proposed a hybrid algorithm to classify blood cells ,especially Acute lymphoblastic leukemia cells (ALL) , if they have leukemia or normal. The Elephant Herd Optimization (EHO) algorithm trains the FF-neural network to reduce the error rate of classification. Elephant Herd Optimization (EHO) algorithm