Current trust models enable decision support at an implicit level by means of thresholds or constraint satisfiability. Decision support is mostly included only for a single binary...
We present a Dynamic Data Driven Application System (DDDAS) to track 2D shapes across large pose variations by learning non-linear shape manifold as overlapping, piecewise linear s...
Malware analysis is the process of determining the behavior and purpose of a given malware sample (such as a virus, worm, or Trojan horse). This process is a necessary step to be a...
Abstract. Cluster ensembles are deemed to be better than single clustering algorithms for discovering complex or noisy structures in data. Various heuristics for constructing such ...
Abstract. We propose and discuss foundations for programmable overlay networks and overlay computing systems. Such overlays are built over a large number of distributed computation...