A Model for Composing Semantic Relations

This paper presents a model to compose semantic relations. The model is independent of any particular set of relations and uses an extended definition for semantic relations. This extended definition includes restrictions on the domain and range of relations and utilizes semantic primitives to characterize them. Primitives capture elementary properties between the arguments of a relation. An algebra for composing semantic primitives is used to automatically identify the resulting relation of composing a pair of compatible relations. Inference axioms are obtained. Axioms take as input a pair of semantic relations and output a new, previously ignored relation. The usefulness of this proposed model is shown using PropBank relations. Eight inference axioms are obtained and their accuracy and productivity are evaluated. The model offers an unsupervised way of accurately extracting additional semantics from text.

[1]  Christiane Fellbaum,et al.  Knowledge Processing On An Extended Wordnet , 1998 .

[2]  Ruth Palombo Weiss.  Memory and Learning. , 2000 .

[3]  Hermann Helbig,et al.  Knowledge Representation and the Semantics of Natural Language , 2005, Cognitive Technologies.

[4]  Xavier Carreras,et al.  Introduction to the CoNLL-2005 Shared Task: Semantic Role Labeling , 2005, CoNLL.

[5]  Tan,et al.  Interactive semantic analysis of Clause-Level Relationships , 1995 .

[6]  Preslav Nakov,et al.  SemEval-2007 Task 04: Classification of Semantic Relations between Nominals , 2007, Fourth International Workshop on Semantic Evaluations (SemEval-2007).

[7]  Dan Moldovan,et al.  Models for the Semantic Classification of Noun Phrases , 2004, HLT-NAACL 2004.

[8]  Preslav Nakov,et al.  SemEval-2010 Task 8: Multi-Way Classification of Semantic Relations Between Pairs of Nominals , 2009, SEW@NAACL-HLT.

[9]  Paul Nulty.  Semantic Classification of Noun Phrases Using Web Counts and Learning Algorithms , 2007, ACL.

[10]  Michael N. Huhns,et al.  Plausible Inferencing Using Extended Composition , 1989, IJCAI.

[11]  Daniel Gildea,et al.  The Proposition Bank: An Annotated Corpus of Semantic Roles , 2005, CL.

[12]  Douglas Herrmann,et al.  A Taxonomy of Part-Whole Relations , 1987, Cogn. Sci..

[13]  Paul R. Cohen,et al.  Beyond ISA: Structures for Plausible Inference In Semantic Networks , 1988, AAAI.

[14]  James H. Martin,et al.  Learning Semantic Links from a Corpus of Parallel Temporal and Causal Relations , 2008, ACL.

[15]  Hermann Helbig.  Knowledge Representation and the Semantics of Natural Language (Cognitive Technologies) , 2005 .

[16]  Sanda M. Harabagiu,et al.  Knowledge processing on an extended wordnet , 1998 .

[17]  Xavier Carreras,et al.  Semantic Role Labeling: An Introduction to the Special Issue , 2008, Computational Linguistics.

[18]  Dan I. Moldovan,et al.  Automatic Discovery of Part-Whole Relations , 2006, CL.

[19]  Dan Roth,et al.  Extraction of Entailed Semantic Relations Through Syntax-Based Comma Resolution , 2008, ACL.

[20]  Du-Seong Chang,et al.  Incremental cue phrase learning and bootstrapping method for causality extraction using cue phrase and word pair probabilities , 2006, Inf. Process. Manag..