We examine the viability of building large polarity lexicons semi-automatically from the web. We begin by describing a graph propagation framework inspired by previous work on con...
Leonid Velikovich, Sasha Blair-Goldensohn, Kerry H...
In this paper we study simple families of clustered graphs that are highly unconnected. We start by studying 3-cluster cycles, which are clustered graphs such that the underlying ...
Pier Francesco Cortese, Giuseppe Di Battista, Maur...
The application of social network analysis techniques, along with graph visualization and interaction, for navigating syndicated web content, a.k.a webfeeds, is presented. Within ...
In this paper, we propose a new methodology based on directed graphs and the TextRank algorithm to automatically induce general-specific noun relations from web corpora frequency c...
We propose a new graph-based semisupervised learning (SSL) algorithm and demonstrate its application to document categorization. Each document is represented by a vertex within a ...