Clustering and Traversals in Concept Association Networks


We view association of concepts as a complex network and analyse its structure. We observe that concept association network is scale-free and has small-world properties. We also study two large scale properties of these networks --- clusters and paths. First, we present an algorithm for clustering these networks which generate qualitatively better clusters than those generated by spectral clustering, a conventional mechanism used for graph partitioning. Next, we study paths generated by human traversals on these networks and contrast it with random walks and shortest distance paths. Our results are a first step towards viewing human cognitive abilities in the light of complex network analysis.

Back to Table of Contents