Cart (Loading....) | Create Account
Close category search window
 

How to Forget the Second Side of the Story: A New Method for the One-Mode Projection of Bipartite Graphs

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Zweig, K.A. ; Interdiscipl. Center for Sci. Comput. (IWR), Univ. of Heidelberg, Heidelberg, Germany

Many relationships naturally come in a bipartite setting: authors that write articles, proteins that interact with genes, or customers that buy, rent or rate products. Often we are interested in the clustering behavior of one side of the graph, i.e., in finding groups of similar articles or products. To find these clusters, a one-mode projection is classically applied, which results in a normal graph that can then be clustered by various methods. For data with strongly skewed degree distributions, a classical one-mode projection leads to very dense graphs with little information. In this article we propose a new method for a meaningful one-mode projection of any kind of bipartite graph B to a sparse general graph G, using a modified version of the so-called leverage. We provide ample experimental evidence that the method creates edges in G only between statistically significant neighbors and that the results are reliable and stable. For this, we present an output sensitive algorithm to compute Kendall's τ. Moreover, for a subset of films in the Netflix prize data set, we can prove that the proposed method not only detects the statistically significantly co-rented films in the data set but that these are also the films that are the most similar ones by content. Thus, our method cannot only be used for the one-mode projection of bipartite graphs in general but also especially for any kind of market basket data to find pairs of most similar products as needed for, e.g., recommendation systems.

Published in:

Advances in Social Networks Analysis and Mining (ASONAM), 2010 International Conference on

Date of Conference:

9-11 Aug. 2010

Need Help?


IEEE Advancing Technology for Humanity About IEEE Xplore | Contact | Help | Terms of Use | Nondiscrimination Policy | Site Map | Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest professional association for the advancement of technology.
© Copyright 2014 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.