Learning to generate CGs from domain specific sentences. LNCS, 2120:44–57, 2001. you find a list of the most recent technical reports of the Max-Planck-Institut für Informatik. They are available by anonymous ftp fromftp.mpi-sb.mpg.de under the directory (2001) [3 citations — 0 self]
Abstract:
Abstract. Automatically generating Conceptual Graphs (CGs) [1] from natural language sentences is a difficult task in using CG as a semantic (knowledge) representation language for natural language information source. However, up to now only few approaches have been proposed for this task and most of them either are highly dependent on one domain or use many manually constructed generation rules. In this paper, we propose a machine-learning based approach that can be trained for different domains and requires almost no manual rules. We adopt a unique grammar system – Link Grammar [2] – for this purpose. The link structures of the grammar are more similar to conceptual graphs than traditional parse trees. Based on the link structure, through the word-conceptualization, concept-folding, link-folding and relationalization processes, we can train the system to generate conceptual graphs from domain specific sentences. An implementation system is currently under development with IBM China Research Lab. 1
