The Needles-in-Haystack Problem

We consider a new data mining problem of detecting the members of a rare class of data, the needles, that have been hidden in a set of records, the haystack. Besides the haystack, a single instance of a needle is given. It is assumed that members of the needle class are similar according to an unknown needle characterization. The goal is to find the needle records hidden in the haystack. This paper describes an algorithm for that task and applies it to several test cases. Key words: Logic, Classification, Feature Selection

• Type: Conference
• Year: 2009
• Where: MLDM
• Authors: Katherine Moreland, Klaus Truemper
• Keywords: Data Mining Problem, Machine Learning, MLDM 2009, Needle Class, Unknown Needle Characterization