Support vector machine (SVM) is a popular supervised learning algorithm based on margin maximization. It has a high training cost and does not scale well to a large number of data points. We propose a multiresolution algorithm MRH-SVM that trains SVM on a hierarchical data aggregation structure, which also serves as a common data input to other learning algorithms. The proposed algorithm learns SVM models using high-level data aggregates and only visits data aggregates at more detailed levels where support vectors reside. In addition to performance improvements, the algorithm has advantages such as the ability to handle data streams and datasets with imbalanced classes. Experimental results show significant performance improvements in compa

This paper sheds the light on the vital role that fractional ordinary differential equations(FrODEs) play in the mathematical modeling and in real life, particularly in the physical conditions. Furthermore, if the problem is handled directly by using numerical method, it is a far more powerful and efficient numerical method in terms of computational time, number of function evaluations, and precision. In this paper, we concentrate on the derivation of the direct numerical methods for solving fifth-order FrODEs  in one, two, and three stages. Additionally, it is important to note that the RKM-numerical methods with two- and three-stages for solving fifth-order ODEs are convenient, for solving class's fifth-order FrODEs. Numerical exa

The experiences in the life are considered important for many fields, such as industry, medical and others. In literature, researchers are focused on flexible lifetime distribution.

In this paper, some Bayesian estimators for the unknown scale parameter  of Inverse Rayleigh Distribution have been obtained, of different two loss functions, represented by Suggested and Generalized loss function based on Non-Informative prior using Jeffery's and informative prior represented by Exponential distribution. The performance of   estimators is compared empirically with Maximum Likelihood estimator, Using Monte Carlo Simulation depending on the Mean Square Error (MSE). Generally, the preference of Bayesian method of Suggeste

The present work provides theoretical investigation of laser photoacoustic one dimensional imaging to detect a blood vessel or tumor embedded within normal tissue. The key task in photoacoustic imaging is to have acoustic signal that help to determine the size and location of the target object inside normal tissue. The analytical simulation used a spherical wave model representing target object (blood vessel or tumor) inside normal tissue. A computer program in MATLAB environment has been written to realize this simulation. This model generates time resolved acoustic wave signal that include both expansion and contraction parts of the wave. The photoacoustic signal from the target object is simulated for a range of laser pulse duration 1

The simulation of passively Q-switching is four non – linear first order differential equations. The optimization of passively Q-switching simulation was carried out using the constrained Rosenbrock technique. The maximization option in this technique was utilized to the fourth equation as an objective function; the parameters, γa, γc and β as were dealt with as decision variables. A FORTRAN program was written to determine the optimum values of the decision variables through the simulation of the four coupled equations, for ruby laser Q–switched by Dy +2: CaF2.For different Dy +2:CaF2 molecules number, the values of decision variables was predicted using our written program. The relaxation time of Dy +2: CaF2, used with ruby was