High assurance in embedded system software is difficult to attain. Verification relies on testing. The unreliable and costly testing process is made much worse because the softwar...
— We present a hybrid data mining approach to detect malicious executables. In this approach we identify important features of the malicious and benign executables. These feature...
Mohammad M. Masud, Latifur Khan, Bhavani M. Thurai...
Mole pattern changes are important cues in detecting melanoma at an early stage. As a first step to automatically register mole pattern changes from skin images, this paper prese...
Supervised learning is a classic data mining problem where one wishes to be be able to predict an output value associated with a particular input vector. We present a new twist on...
David R. Musicant, Janara M. Christensen, Jamie F....
Frequent failures are becoming a serious concern to the community of high-end computing, especially when the applications and the underlying systems rapidly grow in size and compl...