Showing papers in "Journal of Computational Science in 2018"

TL;DR: This improved convolutional neural network combined the parametric rectified linear unit (PReLU) and dropout techniques and is superior to four state-of-the-art approaches.

TL;DR: Empirical analysis demonstrates that forecasting accuracy of the proposed model based on granular intervals is better than non-granular intervals and can take far better decision with the M-factors time series data sets.

TL;DR: Deep learning can provide better detection results than traditional artificial intelligence methods in breast cancer detection, and the combination of parametric ReLU and rank-based stochastic pooling performed the best, with sensitivity, specificity, precision, and accuracy.

TL;DR: A hybrid algorithm namely HAFA, which incorporates certain aspects of firefly optimization and ant colony system algorithms for solving a class of vehicle routing problems, demonstrates the superiority of proposed approach over other existing FA based approaches for solving such type of discrete optimization problems.

TL;DR: This research uses the multi-criteria ratings in developing a new recommendation method for hotel recommendations in e-tourism platforms and uses supervised and unsupervised machine learning techniques to analysis the customers’ online reviews.

TL;DR: In this work, the parallel multi-physics phase-field framework Pace3D is introduced and the general structure of the solver, its modules and the parallelization are described, and various optimization techniques are outlined for increasing the performance of the implemented phase- field models.

TL;DR: In the proposed CAD system, pre-processing is performed to suppress the noise in the mammographic image, then segmentation locates the tumor in mammograms using the cascading of Fuzzy C-Means and region-growing algorithm called FCMRG.

TL;DR: The GWO-ABC is applied to design an optimal fractional order PID (FOPID) controllers for variety of typical benchmark complex transfer functions and trajectory tracking problem of 2 degree-of-freedom (DOF) robotic manipulator and established as viable alternative to design a controller with optimal parameters and enhance the performance of complex systems.

TL;DR: In this work, a classification methodology is proposed to classify FEEG and NFEEG, and it was found that Non Nested Generalized Exemplers (NNge) classifier gave the highest classification accuracy of 99, 99% and specificity of 99%, which is good comparing to proposed methods in the literature.

TL;DR: The proposed path planning method is proposed based on the adaptive polymorphic ant colony algorithm, which achieves superior performance in path planning for smart wheelchairs and is even racing ahead of other state-of-the-art solutions.

TL;DR: To improve the gesture recognition accuracy without training the machine learning algorithm subject specifically, the time-domain EMG features are normalized to the area under the averaged root mean square curve (AUC-RMS).

TL;DR: A new efficient version of recently proposed Antlion Optimizer (ALO) namely Opposition based Levy Flight Antlions optimizer (OB-LF-ALO), conceptualized on the theory of Opposition based learning integrated with Levy Flight for random walk in place of uniform distributed randomWalk in original ALO is proposed.

TL;DR: A computationally intelligent and efficient method based on particle swarm optimization (PSO) is developed that utilizes the features extracted from texture and wavelet domain to select informative wavelet sub-band in social networks.

TL;DR: The overall accuracy offered by neutrosophic sets is accurate, less time consuming and less sensitive to noise and performs well on non-uniform CT images.

TL;DR: A novel hybrid recommendation algorithm is proposed to resolve the two issues by bridging the movie feature and user interest, which aims to make the recommendation results adapt to the change of user interest.

TL;DR: This study is the first among other research studies on curb and on-street parking to consider driver behaviour such as patience, perceptivity and the use of indicators (turn signals), and it also considers the impact of vacant parking-space marking.

TL;DR: The proposed HRS-CE algorithm generates the user profiles that depict the type of content in which a particular user is interested and outperforms the state of the art recommender system models for CS items.

TL;DR: RBIB proposes a novel bagging architecture incorporating a constrained bag creation method, a Risk Induced Bayesian Inference method as a base learner and a cost-sensitive weighted voting combiner, which indicates robustness of the model in handling unseen data without any need for domain specific parameter fine-tuning.

TL;DR: This paper proposes two privacy-preserving mobile application recommendation schemes based on trust evaluation that have sound security, efficiency, accuracy, and robustness and develops two mobile Apps that can be applied in different scenarios, i.e., a centralized cloud service and distributed social networking.

TL;DR: A novel coordination algorithm combining three biologically inspired processes: stigmergy, flocking and evolution is designed, which adapts the swarm coordination according to environmental conditions.

TL;DR: The experimental results show that the proposed recommendation system can expand the time horizon of traffic congestion prediction and further improve the reliability and predictability of decision-making basis for traffic managers and travelers.

TL;DR: The nonlinear linear FEAST algorithm as mentioned in this paper uses complex contour integration to calculate the eigenvectors associated with eigenvalues that are located inside some user-defined region in the complex plane.

TL;DR: A fire code has been developed for the purpose of modelling wildland fires via Large Eddy Simulation (LES) and the use of the level-set approach to track the flame front and information including temperature field, toxic volatiles and soot particle concentrations can be realised in comparison to empirical fire spread models.

TL;DR: Numerical performance of rhoCentralFoam, a compressible solver in OpenFOAM, was studied by modeling round supersonic free jets with varying nozzle exit conditions through Reynolds-Averaged Navier-Stokes approach and good overall agreements can be observed.

TL;DR: The development stages of a small-scale fire whirl including the ignition, flame-rising and fully-developed whirling were successfully captured by a fire field model and the radiation heat feedback towards the liquid fuel was aptly predicted.

TL;DR: The proposed method improves the accuracy of community detection in comparison with other available methods, and the modularity and Normalized Mutual Information and also F1AVG of this proposed method is considerably higher than the earlier available methods in most datasets.

TL;DR: Cognitive parallel programming assistant (PAPA) is presented that aims at educating and assisting novice parallel programmers to avoid common OpenMP mistakes and is evaluated using a user experience questionnaire with a number of novice Parallel programmers at Linnaeus University.

TL;DR: Bat algorithm and Particle Swarm Optimization are used to favor that metric which provides high-quality recommendations for a particular user and results indicate that the MAE while using a combination of weighted trust is much better compared to using a single metric.

TL;DR: A computational framework for scale-bridging combined with an algorithm to automatically and adaptively replace at-scale models within a multiscale model hierarchy with surrogate models in order to reduce the computational cost of multiscales simulations is introduced.

TL;DR: A comparative study has been incorporated that depicts the proposed algorithm successfully minimizes the sparsity problem and the experimental outcome rendering ACCF provides 37%, 59%, 53% more Coverage, Precision and F-Measure than the user-based collaborative filtering even applicable for a small sample of data.