The Future of Human Activity Recognition: Deep Learning or Feature Engineering?

Neural Processing Letters - A significant gap exists in our knowledge of how domain-specific feature extraction compares to unsupervised feature learning in the latent space of a deep neural...     

A Grid-Insensitive LDA Method on Triangular Grids Solving the System of Euler Equations

The performance of the classic upwind-type residual distribution (RD) methods on skewed triangular grids are rigorously investigated in this paper. Based on an improved signals distribution, an improved second order RD method based on the LDA approach is proposed to faithfully replicate the flow physics on skewed triangular grids. It will be mathematically and numerically shown that the improved LDA method is found to have minimal accuracy variations when grids are skewed compared to classic RD and cell vertex finite volume methods on scalar equations and system of Euler equations.     

An energy-efficient cuckoo search algorithm for virtual machine placement in cloud computing data centers

The Journal of Supercomputing - The advent of virtualization technology has created a huge potential application for cloud computing. In virtualization, a large hardware resource is often broken...     

Exploring Students Engagement Towards the Learning Management System (LMS) Using Learning Analytics

Learning analytics is a rapidly evolving research discipline that uses the insights generated from data analysis to support learners as well as optimize both the learning process and environment. This paper studied students’ engagement level of the Learning Management System (LMS) via a learning analytics tool, student’s approach in managing their studies and possible learning analytic methods to analyze student data. Moreover, extensive systematic literature review (SLR) was employed for the selection, sorting and exclusion of articles from diverse renowned sources. The findings show that most of the engagement in LMS are driven by educators. Additionally, we have discussed the factors in LMS, causes of low engagement and ways of increasing engagement factors via the Learning Analytics approach. Nevertheless, apart from recognizing the Learning Analytics approach as being a successful method and technique for analyzing the LMS data, this research further highlighted the possibility of merging the learning analytics technique with the LMS engagement in every institution as being a direction for future research.     

Effect of axial conduction on the ice crystal growth in laminar falling filmsEffet de la conduction axiale sur la vitesse de formation de cristaux de glace à l'intérieur de films tombants laminaires

A finite volume numerical code has been developed to numerically approximate the rate of ice crystal growth in a laminar falling film flowing down a cooled vertical plate. The governing energy equation contains the phase energy as the source term. Enhancement of heat transfer as a result of suspended ice crystals is accounted for in the use of effective values of thermal conductivity, viscosity, thermal diffusivity, and specific heat as function of volumetric concentration of ice crystals in the falling film. Nusselt number, overall heat transfer coefficients between the fluid and cooled plate, and ice crystal growth rate were calculated for different film thicknesses with and without axial diffusion. Nusselt number and ice crystal growth rates were found to be dependent on film thickness. Axial diffusion effects were found to be negligible for larger film thickness (large flowrate).     

Speaker identification in emotional talking environments based on CSPHMM2s

Speaker recognition systems perform almost ideal in neutral talking environments; however, these systems perform poorly in emotional talking environments. This research is devoted to enhancing the low performance of text-independent and emotion-dependent speaker identification in emotional talking environments based on employing Second-Order Circular Suprasegmental Hidden Markov Models (CSPHMM2s) as classifiers. This work has been tested on our speech database which is composed of 50 speakers talking in six different emotional states. These states are neutral, angry, sad, happy, disgust, and fear. Our results show that the average speaker identification performance in these talking environments based on CSPHMM2s is 81.50% with an improvement rate of 5.61%, 3.39%, and 3.06% compared, respectively, to First-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM1s), Second-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM2s), and First-Order Circular Suprasegmental Hidden Markov Models (CSPHMM1s). Our results based on subjective evaluation by human judges fall within 2.26% of those obtained based on CSPHMM2s.     

An assessment of the effectiveness of a rotation forest ensemble for land-use and land-cover mapping

Increasing the accuracy of thematic maps produced through the process of image classification has been a hot topic in remote sensing. For this aim, various strategies, classifiers, improvements, and their combinations have been suggested in the literature. Ensembles that combine the prediction of individual classifiers with weights based on the estimated prediction accuracies are strategies aiming to improve the classifier performances. One of the recently introduced ensembles is the rotation forest, which is based on the idea of building accurate and diverse classifiers by applying feature extraction to the training sets and then reconstructing new training sets for each classifier. In this study, the effectiveness of the rotation forest was investigated for decision trees in land-use and land-cover (LULC) mapping, and its performance was compared with performances of the six most widely used ensemble methods. The results were verified for the effectiveness of the rotation forest ensemble as it produced the highest classification accuracies for the selected satellite data. When the statistical significance of differences in performances was analysed using McNemar's tests based on normal and chi-squared distributions, it was found that the rotation forest method outperformed the bagging, Diverse Ensemble Creation by Oppositional Relabelling of Artificial Training Examples (DECORATE), and random subspace methods, whereas the performance differences with the other ensembles were statistically insignificant.     

Nonlinear generalised dynamic inversion for aircraft manoeuvring control

The nonlinear control problem of aircraft trajectory tracking is tackled in the framework of multiple linear time-varying constrained control using the newly developed paradigm of generalised dynamic inversion. The time differential forms of the multiple constraints encapsulate the control objectives, and are inverted to obtain the reference trajectory-realising control law. The inversion process utilises the Moore–Penrose generalised inverse and the associated nullspace projection, and it predictably involves the problematic generalised inversion singularity. Thus, a singularity avoidance scheme based on a new type of dynamically scaled generalised inverses is introduced that guarantees both asymptotically stable tracking and singularity avoidance. The steady-state closed-loop system allows for two inherently noninterfering control actions working towards a unified goal to exploit the aircraft's control authority over the entire state space. One control action is performed by the particular part of the control law on the range space of the transposed constraint matrix, and it works to impose the prescribed aircraft constrained dynamics. The other control action is performed by the auxiliary part of the control law on the complementary orthogonal nullspace of the constraint matrix, and it provides aircraft's global inner stability using the concept of perturbed feedback linearisation. Numerical simulations of an aggressive multiaxial aircraft coordinated manoeuvre verify the efficacy of designing nonlinear flight control systems via this methodology.     

A steady-state genetic algorithm for multi-product supply chain network design

Supply chain network (SCN) design is to provide an optimal platform for efficient and effective supply chain management (SCM). The problem is often an important and strategic operations management problem in SCM. The design task involves the choice of facilities (plants and distribution centers (DCs)) to be opened and the distribution network design to satisfy the customer demand with minimum cost. This paper presents a solution procedure based on steady-state genetic algorithms (ssGA) with a new encoding structure for the design of a single-source, multi-product, multi-stage SCN. The effectiveness of the ssGA has been investigated by comparing its results with those obtained by CPLEX, Lagrangean heuristic, hyrid GA and simulated annealing on a set of SCN design problems with different sizes.