Solving the stochastic support vector regression with probabilistic constraints by a high-performance neural network model

This paper offers a recurrent neural network to support vector machine (SVM) learning in stochastic support vector regression with probabilistic constraints. The SVM is first converted into an equivalent quadratic programming (QP) formulation in linear and nonlinear cases. An artificial neural network for SVM learning is then proposed. The presented neural network framework guarantees obtaining the optimal solution of the SVM problem. The existence and convergence of the trajectories of the network are studied. The Lyapunov stability for the considered neural network is also shown. The efficiency of the proposed method is shown by three illustrative examples.

     

High-speed GPU implementation of a secret sharing scheme based on cellular automata

The Journal of Supercomputing - Parallel implementation provides a solution for the problem of accelerating cellular automata (CA)-based secret sharing schemes and make them appropriate for bulk...     

OSPM: A design methodology for open strategic planning

This study employs a design science perspective to propose a methodology for open strategic planning (OSP). Habermas’ discourse theory and Bryson’s strategy change cycle are used as informing kernel theories. A methodology is proposed to satisfy the requirements retrieved from the kernel theories. The proposed methodology contains modules for a planning system and a planning process. Design principles are explained through a blueprint of the system and process. The proposed methodology is applied and evaluated in two cases. Contributions to the literature involve extending the literature on OSP to an applicable methodology with guidelines on how to implement open strategy.     

Dynamic modeling and CPG-based trajectory generation for a masticatory rehab robot

Human mastication is a complex and rhythmic biomechanical process which is regulated by a brain stem central pattern generator (CPG). Masticatory patterns, frequency and amplitude of mastication are different from person to person and significantly depend on food properties. The central nervous system controls the activity of muscles to produce smooth transitions between different movements. Therefore, to rehab human mandibular system, there is a real need to use the concept of CPG for development of a new methodology in jaw exercises and to help jaw movements recovery. This paper proposes a novel method for real-time trajectory generation of a mastication rehab robot. The proposed method combines several methods and concepts including kinematics, dynamics, trajectory generation and CPG. The purpose of this article is to provide a methodology to enable physiotherapists to perform the human jaw rehabilitation. In this paper, the robotic setup includes two Gough–Stewart platforms. The first platform is used as the rehab robot, while the second one is used to model the human jaw system. Once the modeling is completed, the second robot will be replaced by an actual patient for the selected physiotherapy. Gibbs–Appell’s formulation is used to obtain the dynamics equations of the rehab robot. Then, a method based on the Fourier series is employed to tune parameters of the CPG. It is shown that changes in leg lengths, due to the online changes of the mastication parameters, occur in a smooth and continuous manner. The key feature of the proposed method, when applied to human mastication, is its ability to adapt to the environment and change the chewing pattern in real-time parameters, such as amplitudes as well as jaw movements velocity during mastication.     

Could-based vehicular networks: a taxonomy,survey, and conceptual hybrid architecture

Wireless Networks - In recent times, vehicular network research has attracted the attention of both researchers and the industry partly due to its potential applications in efficient traffic...     

Correction to: Cloud-based vehicular networks: a taxonomy,survey, and conceptual hybrid architecture

Wireless Networks - The original version of this article unfortunately contained a mistake in the title. The correct title has been published with this erratum.     

On the analysis of fracture mechanisms and mechanical behavior of AA5083-based tri-modal composites reinforced with 5 wt.%B4C and toughened by AA5083 and AA2024 coarse grain phases

In this article, the influence of AA2024 and AA5083 coarse grains on mechanical properties and failure mechanisms of AA5083-5wt. %B₄C tri-modal composite has been discussed. AA2024 and AA5083 powders (<100 µm) were added to mechanically milled AA5083-5 wt.%B₄C powders in 25 and 50 wt.% and the mixtures were consolidated using the hot press and hot extrusion techniques. Results indicated that by adding AA2024 and AA5083 powders as coarse grains, hardness and tensile strength of AA5083-5 wt.%B₄C composite decreased but ductility increased. Moreover, by adding AA2024 powders as coarse grains, fracture mode changed and cracks tended to grow through along AA2024/AA5083-5 wt.%B₄C interface rather than being arrested or deflected. It seemed that dislocation mobility and the interface between coarse grains and ultra-fine grains had the main role in determining the mechanical properties and failure mechanisms in tri-modal AA5083-B₄C composites.     

Using the red-near infrared spectral to estimate ground cover based on vegetative indices

Rapid and accurate estimation of Ground Cover (GC) at regional and global scales for agricultural management application is only possible by using remote sensing (RS). In this study, two Vegetation Indices (VIs) including the Perpendicular Vegetation Index (PVI) and Normalized Difference Vegetation Index (NDVI) were used for estimating GC. Since the parameters of the bare soil line have an important role in calculating GC based on PVI, this line was extracted based on the red-NIRmin (minimum near infrared) method with different intervals (0.0001, 0.0005, and 0.0010). In addition to traditional statistics such as Root Mean Square Error (RMSE), the sensitivity analysis (S) was also used to sharpen the accuracy of the models' estimations. The results indicated that the PVI-based method, in contrast to the NDVI-based approach, had a better performance in estimating GC of wheat. The highest correlation between the observed GC and the estimated GC based on PVI method was achieved in interval length of 0.0005 (R2 = 0.91) with RMSE equal to 8.82. This regression line (GCEST = -3.47 + 0.96 GCOBS) was not significantly different from the 1:1 line. As expected, the best estimation was achieved when the sensitivity of estimated GC based on PVI (length of the interval: 0.0005) was almost constant and low compared to the other models.     

Application of Fuzzy c-means algorithm for the estimation of Asphaltene precipitation

One of problematic topics in petroleum engineering is Asphaltene precipitation issue which causes problems such as tubing plugging and formation damage due to temperature, pressure and composition changes so the notability of this issue increases. In the present investigation a novel Fuzzy c-means (FCM) algorithm was developed to predict precipitated asphaltene as function of dilution ratio, carbon number of precipitants and temperature for solving the problem. The results showed that this novel approach has great ability to predict precipitated asphaltene in terms of aforementioned parameters. The coefficients of determination (R²) for training and testing steps are calculated as 0.9828 and 0.9387 respectively. This great degree of accuracy expresses that the predicting algorithm has potential to be utilized as software for prediction of asphaltene behavior.