Search

Matrix factorization has demonstrated outstanding performance in machine learning. Recently, graph-based matrix factorization has gained widespread attention. However, graph-based methods are only suitable for handling small amounts of data. This paper proposes a fast semi-supervised learning method using only matrix factorization, which considers both global and local information. By introducing bipartite graphs into symmetric matrix factorization, the technique can handle large datasets effectively. It is worth noting that by utilizing tag information, the proposed symmetric matrix factorization becomes convex and unconstrained, i.e., the non-convex problem min x (1 − x 2) 2 is transformed into a convex problem. This allows it to be optimized quickly using state-of-the-art unconstrained optimization algorithms. The computational complexity of the proposed method is O (n m d) , which is much lower than that of the original symmetric matrix factorization, which is O (n 2 d) , and even lower than that of other anchor-based methods, which is O (n m d + m 2 n + m 3) , where n represents the number of samples, d represents the number of features, and m ≪ n represents the number of anchors. The experimental results on multiple public datasets indicate that the proposed method achieves higher performance in less time. [ABSTRACT FROM AUTHOR]

The Ba[(Zr0.2Ti0.8)1−xCex]O3 ceramics were fabricated by liquid phase coating method and the impacts of cerium addition into the barium zirconate titanate matrix on their microstructure and dielectric properties were investigated. According to the XRD analysis, the Ba[(Zr0.2Ti0.8)1−xCex]O3 ceramics retain a cubic perovskite structure without secondary phases. The SEM images indicate that the average grain size and gain size distribution of Ba[(Zr0.2Ti0.8)1−xCex]O3 ceramics can be adjusted by the cerium content. And the minimum average grain size of 4.38 μm is obtained in Ba[(Zr0.2Ti0.8)0.92Ce0.08]O3 ceramic. Local deformation of the perovskite unit cells induced by the substitution of Ce3+ and Ce4+ for B-site host ions results in the reduction of the dielectric maximum temperature Tm as x increases. The room temperature dielectric constant and maximum dielectric constant decrease initially and then increase with the increase of x. The frequency dispersion of dielectric constant and the diffuseness of phase transition are found in the Ba(Zr0.2Ti0.8)O3 matrix after cerium doping. And the temperature stability of permittivity is optimized dramatically by moderate cerium addition in Ba[(Zr0.2Ti0.8)1−xCex]O3 ceramics, making these dielectrics promising for multilayer ceramic capacitor applications. [ABSTRACT FROM AUTHOR]

In this paper, we build a model of coupled differential equations concerning pressure, temperature density and velocity in H-T-H-P (High Temperature-High Pressure) gas wells according to the conservation of mass, momentum and energy. We present an algorithm-solving model by the fourth-order Runge-Kutta method. Basic data from the Dayi Well, 7100 m deep in China, are used for case history calculations and a sensitivity analysis is done for the model. Gas pressure, temperature, velocity and density along the depth of the well are plotted with different productions, different geothermal gradients and different thermal conductivities, intuitively reflecting gas flow law and the characteristics of heat transfer in formation. The results can provide a dynamic analysis of production for H-T-H-P gas wells. © 2011 Canadian Society for Chemical Engineering [ABSTRACT FROM AUTHOR]