0000000000215001
AUTHOR
Gil Shabat
Using affinity perturbations to detect web traffic anomalies
The initial training phase of machine learning algorithms is usually computationally expensive as it involves the processing of huge matrices. Evolving datasets are challenging from this point of view because changing behavior requires updating the training. We propose a method for updating the training profile efficiently and a sliding window algorithm for online processing of the data in smaller fractions. This assumes the data is modeled by a kernel method that includes spectral decomposition. We demonstrate the algorithm with a web server request log where an actual intrusion attack is known to happen. Updating the kernel dynamically using a sliding window technique, prevents the proble…
Fast Computation by Subdivision of Multidimensional Splines and Their Applications
We present theory and algorithms for fast explicit computations of uni- and multi-dimensional periodic splines of arbitrary order at triadic rational points and of splines of even order at diadic rational points. The algorithms use the forward and the inverse Fast Fourier transform (FFT). The implementation is as fast as FFT computation. The algorithms are based on binary and ternary subdivision of splines. Interpolating and smoothing splines are used for a sample rate convertor such as resolution upsampling of discrete-time signals and digital images and restoration of decimated images that were contaminated by noise. The performance of the rate conversion based spline is compared with the…