Incremental Generation of Plural Descriptions: Similarity and Partitioning

Approaches to plural reference generation emphasise descriptive brevity, but often lack empirical backing. This paper describes a corpus-based study of plural descriptions, and proposes a psycholinguisticallymotivated algorithm for plural reference generation. The descriptive strategy is based on partitioning and incorporates corpusderived heuristics. An exhaustive evaluation shows that the output closely matches human data.

• Type: Conference
• Year: 2007
• Where: EMNLP
• Authors: Albert Gatt, Kees van Deemter
• Keywords: EMNLP 2007, Natural Language Processing, Plural Descriptions, Plural Reference Generation, Reference Generation Emphasise