Distributed Path-Based Inference in Semantic Networks

  • Authors:

  • Chain-Wu Lee;Chun-Hsi Huang;Laurence Tianruo Yang;Sanguthevar Rajasekaran

  • Affiliations:

  • Department of Computer Science and Engineering, State University of New York at Buffalo, Buffalo, NY 14260, USA. lee-d@cse.buffalo.edu;Department of Computer Science and Engineering, University of Connecticut, Storrs, CT 06269, USA. huang@cse.uconn.edu;Department of Computer Science and Engineering, ST. Fransis Xavier University, Antigonish, NS, B2G 2W5 Canada. lyang@stfx.ca;Department of Computer Science and Engineering, University of Connecticut, Storrs, CT 06269, USA. rajasek@cse.uconn.edu

  • Venue:
  • The Journal of Supercomputing
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents the task model, instruction set, reasoning scheme, software infrastructure, as well as the experimental results, of a new distributed semantic network system. Unlike the synchronous and static marker passing algorithm previously used for parallel semantic network design, our system operates asynchronously, supporting knowledge sharing, dynamic load balancing and duplicate checking. To better the performance in distributed environments, the system has two collaborating components: the slave module, which performs task execution; and the host module, which interacts with the user and processes the information for the slaves. Our current implementation focuses on path-based knowledge inferences, using ANSI C and the MPICH-G2 with flex lexical analyzer and the yacc parser generator. Tests of individual components have been performed on a SUN multiprocessor server. The experiments demonstrate promising speedups.