Publications

Consider a Bernoulli-Gaussian vector whose components are Vi = XiBi, with Bi Bernoulli-q and Xi CN(0; 2), i.i.d. across i and mutually independent.

Traditional classification approaches generalize poorly on image classification tasks, because of the high dimensionality of the feature space.

We study the use of support vector machines (SVM's) In classifying e-mail as spam or nonspam by comparing it to three other classification algorithms: Ripper, Rocchio, and boosting decision trees,

In this thesis, the general goal was to explore the impact of the agile methods when a large, distributed software organization is transforming to an agile way of developing software.

Mobile phones and PDAs are converging into mobile lifestyle devices that offer a wide range of applications to end-users.

The paper introduces a methodology to provide push services to dynamic IP addressed-mobile users in 2.5G/3G networks, and minimize the unsuccessful network-requested PDP (packet data protocol) cont

Over the last few years Ethernet technology, and services based on Ethernet technology have emerged as an indispensable component of the broadband networking and telecommunications industry, for ei

AMATISTA is a European IST project whose main target is the development of a fault tolerant (FT) set of tools.

In order to secure the Internet of Things, we have proposed to split connected devices into groups, called communities of devices.

This paper investigates supporting IPv6, the next generation IP in a Teledesic-like low earth orbit (LEO) satellite backbone.