We present a distributed machine learning framework based on support vector machines that allows classification problems to be solved iteratively through parallel update algorithm...
In this paper we present an Artificial Immune System (AIS) based security framework, which prevents a number of serious Denial of Service (DoS) attacks. The proposed security frame...
Real-world databases often contain syntactic and semantic errors, in spite of integrity constraints and other safety measures incorporated into modern DBMSs. We present ERACER, an...
Recent years have witnessed an increasing interest in designing algorithms for querying and analyzing streaming data (i.e., data that is seen only once in a fixed order) with only...
Alin Dobra, Minos N. Garofalakis, Johannes Gehrke,...
We present methods for training high quality object detectors
very quickly. The core contribution is a pair of fast
training algorithms for piece-wise linear classifiers, which
...