This paper presents our work on using the graph structure of Wiktionary for synonym detection. We implement semantic relatedness metrics using both a direct measure of information flow on the graph and a comparison of the list of vertices found to be "close" to a given vertex. Our algorithms, evaluated on ESL 50, TOEFL 80 and RDWP 300 data sets, perform better than or comparable to existing semantic relatedness measures.

Abstract-Graph algorithms have wide applicablity to a variety of domains and are often used on massive datasets. Recent standardization efforts such as the GraphBLAS specify a set of key computational kernels that hardware and software developers can adhere to. Graphulo is a processing framework that enables GraphBLAS kernels in the Apache Accumulo database. In our previous work, we have demonstrated a core Graphulo operation called TableMult that performs large-scale multiplication operations of database tables. In this article, we present the results of scaling the Graphulo engine to larger problems and scalablity when a greater number of resources is used. Specifically, we present two experiments that demonstrate Graphulo scaling performance is linear with the number of available resources. The first experiment demonstrates cluster processing rates through Graphulo's TableMult operator on two large graphs, scaled between 2 17 and 2 19 vertices. The second experiment uses TableMult to extract a random set of rows from a large graph (2 19 nodes) to simulate a cued graph analytic. These benchmarking results are of relevance to Graphulo users who wish to apply Graphulo to their graph problems.

Interactive characters that cohabit a shared space with human partners need to generate and interpret references to elements of the virtual world. Natural language allows for a wide range of phrasings for referring to any particular object, and this variation is thought to reflect not only spatial but also cognitive and linguistic factors. Our study attempts to account for the variability in referring forms found in a set of dialogs of two human partners performing a treasure-hunt task in a virtual world. A decision tree classifier was built that predicts the form of 51% of the referring expressions, compared to a baseline of 39% achieved by a heuristic classifier. The classification algorithm can be used by conversational characters to generate referring expressions of the appropriate form.