Semantic Composition with Quotient Algebras

D. Clarke; R. Lutz; D. Weir

Semantic Composition with Quotient Algebras

D. Clarke, R. Lutz, D. Weir

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

225 Downloads (Pure)

Abstract

We describe an algebraic approach for computing with vector based semantics. The tensor product has been proposed as a method of composition, but has the undesirable property that strings of different length are incomparable. We consider how a quotient algebra of the tensor algebra can allow such comparisons to be made, offering the possibility of data-driven models of semantic composition.

Original language	English
Title of host publication	Procs of Geometric Models of Natural Language Semantics
Subtitle of host publication	GEMS-2010
Pages	38-44
Publication status	Published - 2010

Keywords

linguistics
vector-based semantics
tensor algebra

Access to Document

Semantic composition quotient algebraAccepted author manuscript, 210 KB

Cite this

@inproceedings{8abd3a252df34cbcb0a4baf248deee4f,

title = "Semantic Composition with Quotient Algebras",

abstract = "We describe an algebraic approach for computing with vector based semantics. The tensor product has been proposed as a method of composition, but has the undesirable property that strings of different length are incomparable. We consider how a quotient algebra of the tensor algebra can allow such comparisons to be made, offering the possibility of data-driven models of semantic composition.",

keywords = "linguistics, vector-based semantics, tensor algebra",

author = "D. Clarke and R. Lutz and D. Weir",

year = "2010",

language = "English",

pages = "38--44",

booktitle = "Procs of Geometric Models of Natural Language Semantics",

}

TY - GEN

T1 - Semantic Composition with Quotient Algebras

AU - Clarke, D.

AU - Lutz, R.

AU - Weir, D.

PY - 2010

Y1 - 2010

N2 - We describe an algebraic approach for computing with vector based semantics. The tensor product has been proposed as a method of composition, but has the undesirable property that strings of different length are incomparable. We consider how a quotient algebra of the tensor algebra can allow such comparisons to be made, offering the possibility of data-driven models of semantic composition.

AB - We describe an algebraic approach for computing with vector based semantics. The tensor product has been proposed as a method of composition, but has the undesirable property that strings of different length are incomparable. We consider how a quotient algebra of the tensor algebra can allow such comparisons to be made, offering the possibility of data-driven models of semantic composition.

KW - linguistics

KW - vector-based semantics

KW - tensor algebra

M3 - Conference contribution

SP - 38

EP - 44

BT - Procs of Geometric Models of Natural Language Semantics

ER -

Semantic Composition with Quotient Algebras

Abstract

Keywords

Access to Document

Fingerprint

Cite this