Using aggregation for selecting content when generating referring expressions

Previous algorithms for the generation of referring expressions have been developed specifically for this purpose. Here we introduce an alternative approach based on a fully generic aggregation method also motivated for other generation tasks. We argue that the alternative contributes to a more integrated and uniform approach to content determination in the context of complete noun phrase generation. 1 I n t r o d u c t i o n When generating referring expressions (RE), it is generally considered necessary to provide sufficient information so that the reader/hearer is able to identify the intended referent. A number of broadly related referring expression algorithms have been developed over the past decade based on the natural metaphor of 'ruling out distractors' (Reiter, 1990; Dale and Haddock, 1991; Dale, 1992; Dale and Reiter, 1995; Horacek, 1995). These special purpose algor i thms constitute the 'standard' approach to determining content for RE-generation at this time; they have been developed solely for this purpose and have evolved to meet some specialized problems. In particular, it was found early on that the most ambitious RE goal-that of always providing the maximally concise referring expression necessary for the context ('full brevity ')-is NP-haxd; subsequent work o n RE-generation has therefore a t tempted to steer a course between computational tractability and coverage. One common feature of the favored algorithmic simplifications is their incrementality: potential descriptions are successively refined (usually non-destructively) to produce the final RE, which therefore may or may not be minimal. This is also often motivated on grounds of psychological plausibility. In this paper, we introduce a completely different metaphor for determining RE-content that may be considered in contrast to, or in combination with, previous approaches. The main difference lies in an orientation to the organization of a data set as a whole rather than to individual components as revealed during incremental search. Certain opportunities for concise expression that may otherwise be missed are then effectively isolated. The approach applies results from the previously unrelated generation task of 'aggregation', which is concerned with the grouping together of structurally related information. 2 T h e a g g r e g a t i o n b a s e d m e t a p h o r Aggregation in generation has hitherto generally consisted of lists of more or less ad hoc, or case-specific rules that group together paxticulax pre-specified configurations (cf. Dalianis and Hovy (1996) and Shaw (1998)); however Bateman et al. (1998) provide a more rigorous and generic foundation for aggregation by applying results from data-summarization originally developed for multimedia information presentation (Kamps, 1997). Bateman et al. set out a general purpose method for constructing agg r e g a t i o n l a t t i ces which succinctly represent all possible structural aggregations for any given data set. 1 The application of the aggregationbased metaphor to RE-content determination is motivated by the observation that if something is a 'potential distractor' for some intended referent, then it is equally, under appropriate conditions, a candidate for aggregation together with the intended referent. That 1'Structural' aggregation refers to opportunities for grouping inherent in the s tructure of the data and ignoring additional opportunities for grouping that might be found by modifying the data inferentially.