Deep multilayer multiple kernel learning

Multiple kernel learning (MKL) approach has been proposed for kernel methods and has shown high performance for solving some real-world applications. It consists on learning the optimal kernel from one layer of multiple predefined kernels. Unfortunately, this approach is not rich enough to solve relatively complex problems. With the emergence and the success of the deep learning concept, multilayer of multiple kernel learning (MLMKL) methods were inspired by the idea of deep architecture. They are introduced in order to improve the conventional MKL methods. Such architectures tend to learn deep kernel machines by exploring the combinations of multiple kernels in a multilayer structure. However, existing MLMKL methods often have trouble with the optimization of the network for two or more layers. Additionally, they do not always outperform the simplest method of combining multiple kernels (i.e., MKL). In order to improve the effectiveness of MKL approaches, we introduce, in this paper, a novel backpropagation MLMKL framework. Specifically, we propose to optimize the network over an adaptive backpropagation algorithm. We use the gradient ascent method instead of dual objective function, or the estimation of the leave-one-out error. We test our proposed method through a large set of experiments on a variety of benchmark data sets. We have successfully optimized the system over many layers. Empirical results over an extensive set of experiments show that our algorithm achieves high performance compared to the traditional MKL approach and existing MLMKL methods.     

Arabic speech synthesis and diacritic recognition

Text-to-speech system (TTS), known also as speech synthesizer, is one of the important technology in the last years due to the expanding field of applications. Several works on speech synthesizer have been made on English and French, whereas many other languages, including Arabic, have been recently taken into consideration. The area of Arabic speech synthesis has not sufficient progress and it is still in its first stage with a low speech quality. In fact, speech synthesis systems face several problems (e.g. speech quality, articulatory effect, etc.). Different methods were proposed to solve these issues, such as the use of large and different unit sizes. This method is mainly implemented with the concatenative approach to improve the speech quality and several works have proved its effectiveness. This paper presents an efficient Arabic TTS system based on statistical parametric approach and non-uniform units speech synthesis. Our system includes a diacritization engine. Modern Arabic text is written without mention the vowels, called also diacritic marks. Unfortunately, these marks are very important to define the right pronunciation of the text which explains the incorporation of the diacritization engine to our system. In this work, we propose a simple approach based on deep neural networks. Deep neural networks are trained to directly predict the diacritic marks and to predict the spectral and prosodic parameters. Furthermore, we propose a new simple stacked neural network approach to improve the accuracy of the acoustic models. Experimental results show that our diacritization system allows the generation of full diacritized text with high precision and our synthesis system produces high-quality speech.     

Preparation of a new ceramic microfiltration membrane from mineral coal fly ash: Application to the treatment of the textile dying effluents

New microfiltration membranes from mineral coal fly-ash material are obtained using ceramic method. Paste from mineral coal fly ash (obtained by calcinations at 800 °C of non-grinded mineral coal) is extruded to elaborate a porous tubular configuration used as supports. The support heated at 1125 °C, shows an average pore diameter and porosity of about 4.5 μm and 51%, respectively. The properties in terms of mechanical and corrosion resistances are very interesting. The elaboration of the layer based on fly-ash powder (obtained by sintering at 700 °C of a finely grinded mineral coal) is performed by slip-casting method. The heating treatment at 800 °C leads to an average pore size of 0.25 μm. The water permeability determined of this membrane is 475 L/h m² bar. This membrane can be used for crossflow microfiltration. The application to the treatment of the dying effluents generated by the washing baths in the textile industry shows an important decrease of turbidity (inferior to 1 NTU), of chemical oxygen demand (COD) values (retention rate of about 75%) and a total color removal. The performances in term of permeate flux and efficiency were determined and compared to those obtained using a commercial alumina microfiltration membrane. Almost the same stabilised permeate flux was obtained (about 100 L h⁻¹ m⁻²). So, it seems that the prepared membrane is suitable for such wastewater treatment.     

Hydrogen production by methanol aqueous electrolysis using photovoltaic energy: Algerian potential

Hydrogen is considered as the most promising energy carrier for providing a clean, reliable and sustainable energy system. It can be produced from a diverse array of potential feed stocks including water, fossil fuels and organic matter. Electrolysis is the best option for producing hydrogen very quickly and conveniently. Water electrolysis as a source of hydrogen production has recently gained much attention since it can produce high purity hydrogen and can be compatible with renewable energies. Besides the water electrolysis, aqueous methanol electrolysis has been reported in several studies. The aqueous methanol electrolysis proceeds at much lower voltage than that with the water electrolysis. As a result of the substantially lower operating voltage, the energy efficiency for methanol electrolysis can be higher than that for water electrolysis. In this paper, we are interesting to methanol electrolysis in order to produce hydrogen. The relation linking hydrogen production rate to the power needed to electrolyse a unit volume of aqueous methanol solution has been determined. Using this relation, the potential of hydrogen from aqueous methanol solution using a PV solar as the energy system has been evaluated for different locations in Algeria.     

Nanoemulsions: Using emulsifiers from natural sources replacing synthetic ones—A review

In recent years, substantial consideration within the food industry has been aimed at the development of food‐grade nanoemulsions (NE) as promising systems for encapsulating, stabilizing, and delivering bioactive compounds. Although numerous studies have revealed the critical potential of NE, there are still several challenges to overcome them. These include the extensive amounts of synthetic emulsifiers needed for NE formulation, which can potentially be toxic for human health. The interest in safety, and natural emulsifiers have stimulated food manufacturers to develop "label‐friendly" formulations by replacing synthetic emulsifiers with natural alternatives. This review represents a critical and comprehensive summary of the application of natural emulsifiers as potential substitutes for synthetic emulsifiers in NE production, with particular emphasis on the newly identified natural emulsifiers. Some recent reports showed the excellent emulsifying properties of various natural emulsifier extracted from natural resources, to produce NE, and therefore, might be generalized for further industrial applications. Future trends are encouraged to identify novel natural emulsifiers from industrial food by‐products that may demonstrate highly effective emulsifiers.     

Elaboration and characterisation of fly ash based mineral supports for microfiltration and ultrafiltration membranes

The development and the characterisation of a new support for microfiltration and ultrafiltration membranes from fly ash applied to filtration are presented. The choice of this material is based primarily on its low cost (as fly ash is produced abundantly as a solid waste by coal fired power plants in many parts of the world).The support, with tubular configuration, was prepared from fly ash which was obtained by the sintering of non-grinded mineral coal at 800 °C. The resulting powder, mixed with organic additives and water, could be extruded to elaborate a porous structure. The firing temperature of the support is 1125 °C. The morphology of the surface and the cross-section observed on scanning electron microscope (SEM) are homogeneous and do not present any macrodefects (cracks, etc.). The mean pore diameter, measured by mercury porosimetry is 4.5 μm and the pore volume is 51%. The tubes display a good mechanical and chemical resistance that allow to use them as supports for tangential micro- and ultrafiltration processes.     

Security analysis on “Three-factor authentication protocol using physical unclonable function for IoV”

International Journal of Information Security - The advent of the Internet of Things has enriched the network field with new themes, among which we find the Internet of Vehicles (IoV). IoV improved...     

Detailed manufacturing process of a tubular carbon microfiltration membrane for industrial wastewater treatment

Journal of Porous Materials - This paper discusses the preparation of an asymmetric microfiltration carbon membrane made of mineral coal powder and phenolic resin. Tubular supports were obtained...     

Elaboration of new ceramic microfiltration membranes from mineral coal fly ash applied to waste water treatment

This work aims to develop a new mineral porous tubular membrane based on mineral coal fly ash. Finely ground mineral coal powder was calcinated at 700 °C for about 3 h. The elaboration of the mesoporous layer was performed by the slip-casting method using a suspension made of the mixture of fly-ash powder, water and polyvinyl alcohol (PVA). The obtained membrane was submitted to a thermal treatment which consists in drying at room temperature for 24 h then a sintering at 800 °C. SEM photographs indicated that the membrane surface was homogeneous and did not present any macrodefects (cracks, etc…). The average pore diameter of the active layer was 0.25 μm and the thickness was around 20 μm. The membrane permeability was 475 l/h m² bar.This membrane was applied to the treatment of the dying effluents generated by the washing baths in the textile industry. The performances in term of permeate flux and efficiency were determined and compared to those obtained using a commercial alumina microfiltration membrane. Almost the same stabilised permeate flux was obtained (about 100 l h⁻¹ m⁻²). The quality of permeate was almost the same with the two membranes: the COD and color removal was 75% and 90% respectively.