Shahab Shamshirband's research while affiliated with University of Kuala Lumpur and other places

... Multiple input neuron b=f (Vp+k) Mobile Information Systems backpropagation to reduce the error. However, it has the weakness of small limitations, requires a lot of structure and learning parameters, and is easy to fall into local optimum [17]. ...

... The results showed that the model has a better generalization ability and practical value in water level prediction. On a deeper level, Soleymani et al. (2016) designed a new method to predict the water level of rivers. Their method utilizes the firefly algorithm (FFA) to train the radial basis function (RBF), and the FFA is used to interpolate the RBF to predict the best solution. ...

... In the beginning, the project's survival was the most critical factor. Because of this, early studies mainly focused on technical or economic assessment [6][7][8][9][10][11][12]. Some studies have begun to take the environmental aspect [13][14][15][16] into consideration as increasingly more people are becoming concerned about the planet's environment. ...

... Malekmohamadi et al. (2011) andJames et al. (2018) applied different ML models to predict wave conditions in coastal waters indicating the efficiency of the employed model. Roy et al. (2016) developed a wave height prediction model, using a novel method based on SVMs coupled with the Firefly Algorithm and compared the predicted results with ANNs and genetic programming. Berbić et al. (2017) applied ANNs and SVMs for the purpose of realterm significant wave forecasting. ...

... The total demand of all the customers in a constructed route should not exceed the total capacity of the vehicles. In some problems, vehicle usage time limitation is considered together with the traveled distance limitation: the distance traveled and the travel time spent on the vehicle for the route must not exceed the maximum allowed limits [9,10]. Minimizing the number of vehicles besides minimizing the distance traveled through the routes, and minimizing total travel time are among the main objectives of this problem. ...

... They concluded that ANFIS performed slightly better than ANN models and was more superior to MNLR and MLR. The classical ANFIS was employed extensively within the different research groups to assess Cd in different grade control structures: labyrinth side-weir, semi-elliptical side weir, Two-Cycle Labyrinth Side Weir, Type B piano-key side weir, etc [6,[76][77][78][79][80][81][82][83][84][85][86]. The first developed ANFIS model, which used an evolutionary algorithm, was employed by Khoshbin et al. [87]. ...

... One of the many artificial intelligence methods are Artificial Neural Network. (Hashim et al. 2015). ...

... In recent years, clustering algorithms have been widely used for meteorological and climatological studies, although the choice of the input parameters represents a critical constraint for successful algorithm setup (see the review by Huth et al. 2008 for details about clustering techniques and meteorological applications). Among these, thanks to its simplicity of implementation and computation efficiency, the k-means algorithm is widely used for the classification of environmental data-e.g., observations from meteorological stations (Li et al. 2020), evapotranspiration measurements (Ferreira et al. 2019), and air quality parameters (Adame et al. 2012)-as well as for wind energy resource assessments (Al-Shammari et al. 2016) and rainfall prediction (Geetha and Nasira 2014). ...

... Literature [34] used extreme random trees to classify daytime weather types into four categories, and based on weather type used multiple machine learning models for power generation prediction, and the results showed that the SVM model outperformed other traditional prediction models when the weather changed frequently. Literature [35] used a SVM model to simplify the complex mathematical problem of output power prediction and combined it with wavelet, radial basis function, and firefly algorithm and verified that the model can obtain more accurate prediction results. Literature [36] proposed a hybrid model combining wavelet transform, particle swarm optimization, and SVM (WT-PSO-SVM), and applied the model to short-term power prediction of real microgrid generation systems. ...

... Since these neural networks capture nonlinearities more effectively than look-up tables, feedforward neural networks were determined to reduce the number of engine measurements required to generate engine-calibration maps [600]. With the increasing popularity of biofuels and fuel additives, these ML algorithms are under investigation as a way to handle the added dimensionality in calibration tables that arises from variations in fuel composition [601,602]. These surrogate models could also be employed to improve other aspects of engine calibration, as highlighted by the use of SVMs to reduce the number of experimental measurements needed to explore novel optimization algorithms for engine calibration [603]. ...