A Generative Model for Self/Non-self Discrimination in Strings

A statistical generative model is presented as an alternative to negative selection in anomaly detection of string data. We extend the probabilistic approach to binary classification from fixed-length binary strings into variable-length strings from a finite symbol alphabet by fitting a mixture model of multinomial distributions for the frequency of adjacent symbols. Robust and localized change analysis of text documents is viewed as an application area.

• Type: Conference
• Year: 2009
• Where: ICANNGA
• Authors: Matti Pöllä
• Keywords: Algorithms, ICANNGA 2009, Statistical Generative Model, finite Symbol Alphabet, fixed-length Binary Strings