by Yanchang Zhao, RDataMining.com
There is an excellent tutorial on outlier detection techniques, presented by Hans-Peter Kriegel et al. at ACM SIGKDD 2010. It presents many popular outlier detection algorithms, most of which were published between mid 1990s and 2010, including
– statistical tests,
– depth-based approaches,
– deviation-based approaches,
– distance-based approaches,
– density-based approaches, and
– high-dimensional approaches.
The slides of the tutorial can be downloaded at http://www.dbs.ifi.lmu.de/~zimek/publications/KDD2010/kdd10-outlier-tutorial.pdf.