Knowledge Discovery from Evolutionary Trees

Abstract

In this chapter we present new techniques for discovering knowledge from evolutionary trees. An evolutionary tree is a rooted unordered labeled tree in which there is a root and the order among siblings is unimportant. The knowledge to be discovered from these trees refers to “cousin pairs” in the trees. A cousin pair is a pair of nodes sharing the same parent, the same grandparent, or the same great-grandparent, etc. Given a tree T, our algorithm finds all interesting cousin pairs of T in O(T 2) time where T is the number of nodes in T. We also extend this algorithm to find interesting cousin pairs in multiple trees. Experimental results on synthetic data and real trees demonstrate the scalability and effectiveness of the proposed algorithms. To show the usefulness of these techniques, we discuss an application of the cousin pairs to evaluate the consensus of equally parsimonious trees and compare them with the widely used clusters in the trees. We also report the implementation status of the system built based on the proposed algorithms, which is fully operational and available on the world-wide web.