Sharif Digital Repository

Multi-label classification has received considerable interest in recent years. Multi-label classifiers usually need to address many issues including: handling large-scale datasets with many instances and a large set of labels, compensating missing label assignments in the training set, considering correlations between labels, as well as exploiting unlabeled data to improve prediction performance. To tackle datasets with a large set of labels, embedding-based methods represent the label assignments in a low dimensional space. Many state-of-the-art embedding-based methods use a linear dimensionality reduction to map the label assignments to a low-dimensional space. However, by doing so, these... 

Multi-label classification has received considerable interest in recent years. Multi-label classifiers usually need to address many issues including: handling large-scale datasets with many instances and a large set of labels, compensating missing label assignments in the training set, considering correlations between labels, as well as exploiting unlabeled data to improve prediction performance. To tackle datasets with a large set of labels, embedding-based methods represent the label assignments in a low-dimensional space. Many state-of-the-art embedding-based methods use a linear dimensionality reduction to map the label assignments to a low-dimensional space. However, by doing so, these... 

SnO2 is considered as one of the high specific capacity anode materials for Lithium-ion batteries. However, the low electrical conductivity of SnO2 limits its applications. This manuscript reports a simple and efficient approach for the synthesis of Sb-doped SnO2 nanowires (NWs) core and carbon shell structure which effectively enhances the electrical conductivity and electrochemical performance of SnO2 nanostructures. Sb doping was performed during the vapor-liquid-solid synthesis of SnO2 NWs in a horizontal furnace. Subsequently, carbon nanolayer was coated on the NWs using the DC Plasma Enhanced Chemical Vapor Deposition approach. The carbon-coated shell improves the Solid-Electrolyte... 

We prepared three ferrite nanocatalysts: (i) copper ferrite (CuFe 2O4) (ii) ferrite where cobalt was substituted by nickel (Nix Co1-x Fe2O4, with x=0, 0.2, 0.4, 0.6), and (iii) ferrite where nickel was substituted by zinc (Zn y Ni1-y Fe2O4 with y=1, 0.7, 0.5, 0.3), by the sol-gel method. The X-ray diffraction patterns show that the ferrite samples have been crystallized in the cubic spinel structural phase. We obtained the size of grains by field emission scanning electron microscopy images and their magnetic properties by vibrating sample magnetometer. Next, carbon nanotubes were grown on these nanocatalysts by the catalytic chemical vapor deposition method. We show that the catalytic... 

Multi-label classification with a large set of labels is a challenging task. Label-Space Dimension Reduction (LSDR) is the most popular approach that addresses this problem. LSDR methods project the high-dimensional label vectors onto a low-dimensional space that can be predicted from the feature space. Many LSDR methods assume that the training data provide complete label vector for all training samples while this assumption is usually violated particularly when label vectors are high dimensional. In this paper, we propose a probabilistic model that has an effective mechanism to handle missing and noisy labels. In the proposed Bayesian network model, a set of auxiliary random variables,... 

Ferrite nanocrystals are an interesting material due to their rich physical properties. Here we add nonmagnetic dopants Zn and Cu to nickel ferrite nanocrystals, Ni1-xMxFe2O4 (0 ≤ x ≤ 1, M = Cu, Zn), and study how relevant properties of the samples are modified accordingly. Basically, these dopings cause a rearrangement of Fe +3 ions into the two preexisting octahedral and tetrahedral sites. In fact, this, we show, induces pertinent magnetic properties of the doped samples. In the case of the Cu-doping, the Jahn-Teller effect also emerges, which we identify through the Fourier Transform Infra-Red Spectroscopy of the samples. Moreover, we show an increase in the lattice parameters of the... 

Adsorption equilibrium data for CO, CO2, CH4 and C2H4 on 5A molecular sieve were obtained at (303, 373 and 573) K and at partial pressures up to 1000 kPa. An accurate and simple volumetric apparatus was used to measure gas sorption. The Langmuir and Sips equations were used to correlate the experimental data. In spite of the relative simplicity of both isotherms, the correlation between experimental and theoretical data is good. The Sips equation with three parameters has expected a better fit than the Langmuir isotherm with two parameters. Based upon the Langmuir equation the enthalpy of adsorption and pre-exponential factor in the van't Hoff correlation were derived for all four gases. ©... 

One of the most attractive features of Simulated Moving Beds (SMBs) is their ability to perform reaction and separation simultaneously in order to increase reaction yields. In this paper, an appropriate configuration for the operation of such systems in gas phase has been presented and to improve the separation efficiency, Pressure Swing Adsorption (PSA) is also employed. Fischer-Tropsch Synthesis (FTS) has been chosen to verify this idea. After preparing the catalyst, choosing appropriate adsorbent and constructing the necessary experimental set up, experiments have been performed to achieve maximum yield of desired products (i.e., ethylene and ethane). Effective variables, such as flow... 

We analyze an experiment in which two distinct superconducting phases YBa2Cu3O7-δ (Y123) and Y 3Ba5Cu8O18-δ (Y358) coexisted. This experiment enabled us to characterize the recently introduced Y358 phase in contrast to the Y123 phase, thus to resolve some discrepancies reported in associated properties of Y358. Specifically, our experiment indicates the transition temperature Tmid C = 105 K and 94 K for Y358 and Y123, respectively, and that Y358 has five CuO2 planes and three CuO chains, with Pmm2 symmetry and lattice parameter (a, b, c) = (3.845, 3.894, 31.093) Å, in agreement with density functional theory predictions for this specific structure  

The bale packings that are used as catalyst containing packings in reactive distillation columns are made and geometrically characterized. Pressure drop and holdup values are measured in a 10.4 cm i.d. glass column packed with bales using an air/water system. The arrangement of the bales in the bed is similar to that of industrial columns. The obtained data are used for evaluating the Stichlmair and the Eckert pressure drop models. Neither of these two models could fit the data properly. When a slight modification is made in the Eckert model, a specific model which fits the data well is prepared. This model also accounts for the effect of geometrical characteristics of the packed bed on... 

The nitrate-citrate gel exhibits auto-catalytic behavior, which can be used to synthesize nanocrystalline YIG powders. In this study, yttrium iron garnet (Y3Fe5O12) nanocrystalline powders were prepared by a sol-gel auto-combustion process. The influence of metal nitrates to citric acid molar ratio (MN/CA) of the precursor solution on the combustion behavior and crystallite size of synthesized powders was investigated by scanning electron microscopy (SEM), thermal analyses (DTA/TGA) and X-ray diffraction (XRD). The results show that with increasing MN/CA value, the combustion rate increases and the single-phase YIG forms at a higher temperature. The crystallite size of the single phase YIG... 

Nanocrystalline Y3-xMMxFe5O12 powders (MM denotes Misch-metal, x = 0.0, 0.25, 0.5, 0.75, and 1.0) were synthesized by a sol-gel combustion method. Magnetic properties and crystalline structures were investigated using X-ray diffraction (XRD), a vibrating sample magnetometer (VSM), and a scanning electron microscope. The XRD patterns showed that the single-phase garnet of Y3-xMMxFe5O12 was formed at x values ≤ 1.0. The saturation magnetization of powders increased with decreasing MM content and reached the maximum value at Y3 Fe5O12. The crystallite size of powders calcined at 800°C for 3 h was in the range of 38-53 nm. © 2008 American Ceramic Society  

The nitrate-citrate gels exhibit auto-catalytic behavior, which can be used to synthesize the nanocrystalline YIG powders. In this study yttrium iron garnet (Y3Fe5O12) nanocrystalline powders were prepared by a sol-gel auto-combustion process. The influence of pH value of the precursor solution on the combustion behavior and the garnet phase formation of synthesized powders were investigated by scanning electron microscopy (SEM), thermal analysis (DTA/TGA), infrared (IR) spectroscopy and X-ray diffraction technique (XRD). The results show that with increasing pH value, the combustion rate increases. The as-burnt powder prepared with pH 1 yielded a single phase YIG after calcination at 800... 

A local radial basis function (RBF) meshless method is applied for solution of the Burgers' equation with different initial and boundary conditions of various complexities. Local-RBF collocation is employed for discretization in space, whilst the unsteady term is handled via a simple explicit time discretization. Moreover, in case of non-smooth initial conditions with high Reynolds numbers, a treatment is proposed for inability of local-RBF methods to solve such problems. The scheme is validated over a variety of benchmark problems and very good agreement is found with existing analytical and numerical solutions  

The objective of this work is to present a finite element formulation for dynamic analysis of sandwich shells with viscoelastic core under large deformation. The present study is based on an incremental updated Lagrangian approach together with the Newmark integration scheme. The viscoelastic constitutive model which is used to define the behavior of the core, comes from the Riesz theorem and the corresponding creep functions are estimated using Dirichlet-Prony series. Also, the viscoelastic deferred strain is derived in an appropriate incremental form using the state variables. The employed layerwise shell element which is based on zig-zag theory has eight nodes on its mid layer. What's... 

In this paper, a numerical model is developed based on the X-FEM technique to simulate the proppant transport and tip screen-out in hydraulic fracturing. The governing equations are based on the momentum balance and mass conservation of the fluid. The hydro-mechanical coupling between the fracture and surrounding porous medium is fulfilled through the weak form of the governing equations. The fluid inflow within the fracture is modeled using the one-dimensional mass conservation of the injected slurry and proppant along the fracture, in which the viscosity of the slurry is dependent on the proppant concentration. The transition from the Poiseuille to Darcy flow regime is incorporated into... 

A semi-analytical thermoelasticity solution for hollow and solid rotating axisymmetric disks made of functionally graded materials is presented. The radial domain is divided into some virtual sub-domains in which the power-law distribution is used for the thermomechanical properties of the constituent components. Imposing the necessary continuity conditions between adjacent sub-domains, together with the global boundary conditions, a set of linear algebraic equations are obtained. Solution of the linear algebraic equations yields the thermoelastic responses for each sub-domain as exponential functions of the radial coordinate. Some results for the stress, strain and displacement components... 

One of the most popular models that has been applied to predict the fluid velocity inside the fracture with impermeable walls is the cubic law. It highlights that the mean flux along the fracture is proportional to the cubic of fracture aperture. However, for a fractured porous medium, the normal and tangential interface conditions between the fracture and porous matrix can change the velocity profile inside the fracture. In this paper, a correction factor is introduced for flow equation along the fracture by imposing the continuity of normal and tangential components of velocity at the interface between the fracture and porous matrix. As a result, the mean velocity inside the fracture... 

A new discretization scheme for the current density in the hydrodynamic model is presented. This scheme is an extension of the Scharfetter-Gummel (S-G) method and is derived without neglecting the convective term in the current density. Simulation of two model n-i-n diodes shows that this method gives better numerical results than the conventional upwind and S-G methods. © 2004 Elsevier B.V. All rights reserved