Incremental Generation of Plural Descriptions: Similarity and Partitioning

Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 102–111, Prague, June 2007. c⃝2007 Association for Computational Linguistics Incremental generation of plural descriptions: Similarity and partitioning Albert Gatt and Kees van Deemter Department of Computing Science University of Aberdeen {agatt,kvdeemte}@csd.abdn.ac.uk Abstract Approaches to plural reference generation emphasise descriptive brevity, but often lack empirical backing. This paper describes a corpus-based study of plural descrip- tions, and proposes a psycholinguistically- motivated algorithm for plural reference generation. The descriptive strategy is based on partitioning and incorporates corpus- derived heuristics. An exhaustive evaluation shows that the output closely matches hu- man data. 1 Introduction Generation of Referring Expressions (GRE) is a well-studied sub-task of microplanning in Natural Language Generation. Most algorithms in this area view GRE as a content determination problem, that is, their emphasis is on the construction of a se- mantic representation which is eventually mapped to a linguistic realisation (i.e. a noun phrase). Con- tent Determination for GRE starts from a Knowledge Base (KB) consisting of a set of entities U and a set of properties P represented as attribute-value pairs, and searches for a description D ⊆ P which distin- guishes a referent r ∈ U from its distractors. Under this view, reference is mainly about identification of an entitiy in a given context (represented by the KB), a well-studied pragmatic function of definite noun phrases in both the psycholinguistic and the compu- tational literature (Olson, 1970). For example, the KB in Table 1 represents 8 en- tities in a 2D visual domain, each with 6 attributes, including their location, represented as a combina- tion of horizontal (X) and vertical (Y) numerical co- TYPE COLOUR ORIENTATION SIZE X Y e1 desk red back small 3 1 e2 sofa blue back small 5 2 e3 desk red back large 1 1 e4 desk red front large 2 3 e5 desk blue right large 2 4 e6 sofa red back large 4 1 e7 sofa red front large 3 3 e8 sofa blue back large 3 2 Table 1: A visual domain ordinates. To refer to an entity an algorithm searches through values of the different attributes. GRE has been dominated by Dale and Reiter’s (1995) Incremental Algorithm (IA), one version of which, generalised to deal with non-disjunctive plural references, is shown in Algorithm 1 (van Deemter, 2002). A non-disjunctive reference to a set R is possible just in case all the elements of R can be distinguished using the same attribute-value pairs. Such a description is equivalent to the logical conjunction of the properties in question. This al- gorithm, IAplur, initialises a description D and a set of distractors C [1.1–1.2], and traverses an ordered list of properties, called the preference order (PO) [1.3], which reflects general or domain-specific pref- Algorithm 1 IAplur(R,U,PO 1: D ← ∅ 2: C ← U − R 4:3: for ⟨A : v⟩ ∈ PO do if R ⊆ [[ ⟨A : v⟩ ]] ∧ [[ …

Content truncated to 500 words