In this paper we present an algorithm of binary classification based on Support Vector Machines. It is combined with a modified Boosting algorithm. It run faster than the original SVM algorithm with a similar generalization error and equal complexity model but it has more compact representation. [ABSTRACT FROM AUTHOR]
The present paper will show that throughout the development of a deterministic algorithm the uniformity of the pressure service level in the Water Distribution Network (WDN) can be maximized. To obtain this in an optimal way, one must maximize an Index that relates the power by unit of weight dissipated by the network and the power by unit of weight available to dissipate by the same one (Resilience Index [2]), it improves the ability of the WDN to cope with possible failures in the system, slows down the pipes renovation processes and diminishes the level of leaks in the network. [ABSTRACT FROM AUTHOR]
This paper is focused on the extraction of myocardial borders (septum, endocardium and epicardium) on M-mode echocardiograms. The proposed method combines two algorithms, each developed specially to extract certain borders. The extraction of the epicardium is done by means of a basic maximum tracking algorithm [8]. For the extraction of the endocardium and the left wall of the interventricular septum, a 2 stages discontinuity detector [6] is used. In the first stage the local mean of the image is calculated, and for the second stage the local variance is used. The output of this detector represents the discontinuities in the M-mode echocardographic image. Additionally, the method allows for manually correction of errors in the detection. Experimental examples are presented to test the performance of the methodology. [ABSTRACT FROM AUTHOR]