Enhancing LOCATION Information Using Semantic Composition

This paper presents a method to enhance location awareness by using semantic composition of AT-LOCATION and PART-WHOLE semantic relations. The method generates axioms that infer new location relations based on relations that are extracted by a semantic parser. Experimental study with WordNet glosses shows that the method increases the amount of location knowledge by two orders of magnitude. The precision of the results is 98%.

[1]  Gaetano Borriello,et al.  Location Systems for Ubiquitous Computing , 2001, Computer.

[2]  George A. Miller,et al.  WordNet: A Lexical Database for English , 1995, HLT.

[3]  Marcus Kracht,et al.  On the Semantics of Locatives , 2002 .

[4]  ADRIANA BADULESCU,et al.  A Semantic Scattering model for the automatic interpretation of English genitives , 2009, Natural Language Engineering.

[5]  Barry Brumitt,et al.  Topological World Modeling Using Semantic Spaces , 2001 .

[6]  Jan Tore Lønning,et al.  A Minimal Recursion Semantic Analysis of Locatives , 2009, Computational Linguistics.

[7]  Junzhong Gu,et al.  Event Based Semantic Location Model in Cooperative Mobile Computing , 2009, 2009 Ninth International Conference on Hybrid Intelligent Systems.

[8]  Salil Pradhan,et al.  Semantic location , 2000, Personal Technologies.

[9]  Ido Dagan,et al.  The Third PASCAL Recognizing Textual Entailment Challenge , 2007, ACL-PASCAL@ACL.