which faithfully reveals the intrinsic neighborhood density. It uses a scaledependent
umbrella operator to bridge global and local properties, which
makes LAD more comprehensive within an adaptive scope of neighborhood.
To offer more stability of local density measurement on scaling parameter
tuning, we formulated Fermi Density Descriptor (FDD) which measures the
probability of a fermion particle being at a specific location, that corresponds
to the neighborhood density. By choosing the stable energy distribution
function, FDD steadily distinguishes anomalies from normal instances with
most of the scaling parameter settings. To further enhance the efficacy of
our proposed algorithms, we explored the utility of Anisotropic Gaussian Kernel
(AGK) which offers better manifold-aware affinity information. We
also quantified and examined the effect of different Laplacian normalizations
for the purpose of anomaly detection. Comprehensive experiments on both
synthetic and benchmark datasets verified that our proposed algorithms outperform
the existing anomaly detection algorithms. This work is recorded in
Chapter 4. The related publications include: