Representing Semantic Graphs in a Self-Organizing Map

A long-standing problem in the field of connectionist language processing has been how to represent detailed linguistic structure. Approaches have ranged from the encoding of syntactic trees in Raam to the use of a mechanism to query meanings in a "gestalt layer". In this article, a technique called semantic self-organization is presented that allows for the optimal allocation and explicit representation of semantic dependency graphs on a Som -based grid. This technique has been successfully used in a connectionist natural language processing architecture called InSomNet to scale up the subsymbolic approach to represent sentences in the LinGO Redwoods HPSG Treebank drawn from the VerbMobil Project and annotated with rich semantic information. InSomNet was also shown to retain the cognitively plausible behavior detailed in psycholinguistics research. Consequently, semantic self-organization holds considerable promise as a basis for real-world natural language understanding systems that mimic human linguistic performance.

[1]  Dan Flickinger,et al.  Minimal Recursion Semantics: An Introduction , 2005 .

[2]  M. Schlesewsky,et al.  Gradience in grammar : generative perspectives , 2006 .

[3]  Marshall R. Mayberry Iii and Risto Miikkulainen Incremental Nonmonotonic Sentence Interpretation through Semantic Self-Organization , 2008 .

[4]  Frank Keller,et al.  Probabilistic Grammars as Models of Gradience in Language Processing , 2006 .

[5]  P MarcusMitchell,et al.  Building a large annotated corpus of English , 1993 .

[6]  Josef Ruppenhofer,et al.  FrameNet II: Extended theory and practice , 2006 .

[7]  Risto Miikkulainen,et al.  Incremental nonmonotonic parsing through semantic self-organization , 2003 .

[8]  Risto Miikkulainen,et al.  Natural Language Processing with Subsymbolic Neural Networks , 2019, Neural Network Perspectives on Cognition and Adaptive Robotics.

[9]  Marc Brysbaert,et al.  Exposure-based models of human parsing: Evidence for the use of coarse-grained (nonlexical) statistical records , 1995 .

[10]  Marshall R. Mayberry,et al.  Using a Sequential SOM to Parse Long-term Dependencies , 2020, Proceedings of the Twenty First Annual Conference of the Cognitive Science Society.

[11]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[12]  Beatrice Santorini,et al.  Building a Large Annotated Corpus of English: The Penn Treebank , 1993, CL.

[13]  Stephan Oepen,et al.  Parse Disambiguation for a Rich HPSG Grammar , 2002 .

[14]  Antony Browne,et al.  Neural Network Perspectives on Cognition and Adaptive Robotics , 1997 .

[15]  D. Kirsh,et al.  Proceedings of the 25th annual conference of the Cognitive Science Society , 2003 .

[16]  Holly P. Branigan,et al.  Proceedings of the 21st Annual Conference of the Cognitive Science Society , 1999 .