To probe network characteristics, two predominant ways of network comparison are global property statistics and subgraph enumeration. However, they suffer from limited information and exhaustible computing. Here, we present an approach to compare networks from the perspective of data analysis. Initially, the approach projects each node of original network as a high-dimensional data point, and the network is seen as clouds of data points. Then the dispersion information of the principal component analysis (PCA) projection of the generated data clouds can be used to distinguish networks. We applied this node projection method to the yeast protein-protein interaction networks and the Internet Autonomous System networks, two types of networks with several similar higher properties. The method can efficiently distinguish one from the other. The identical result of different datasets from independent sources also indicated that the method is a robust and universal framework. © 2009 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.
CITATION STYLE
Li, W., & Yang, J. Y. (2009). Comparing networks from a data analysis perspective. In Lecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering (Vol. 5 LNICST, pp. 1907–1916). https://doi.org/10.1007/978-3-642-02469-6_67
Mendeley helps you to discover research relevant for your work.