xPACS

Explaining Anomalies in Groups with Characterizing Subspace Rules

xPACS can extract succinct anomalous patterns of (i) different types (ii) hidden in arbitrary subspaces of a high dimensional space (iii) interpretable by human analysts

xPACS is accurate in detecting anomalies as well as explaining them.

Code Data Arxiv
mmacha@cmu.edu   ·   lakoglu@andrew.cmu.edu
Website credits   ·   emaad@cmu.edu