A novel representation of graph structures in web mining and data analysis

The paper presents a new graph representation, the graph matrix, which combines the adjacency matrix with the linked lists allowing for the fastest possible access to different types of information on a graph. This is increasingly important for a high search performance, for instance, for rapidly extracting information from the link structure in a hub and authority graph of the World-Wide-Web. A very recent application for the proposed data structure arises from categorical data clustering defining proximity and similarity of data through their patterns of co-occurrence. (C) 2004 Elsevier Ltd. All rights reserved.