Generation of referring expressions : assessing the incremental algorithm

Abstract

A substantial amount of recent work in natural language generation has focussed on the generation of "one-shot" referring expressions whose only aim is to identify a target referent. Dale and Reiter's Incremental Algo- rithm (ia) is often thought to be the best algorithm for maximising the similarity to referring expressions produced by people. We test this hy- pothesis by eliciting referring expressions from human subjects and com- puting the similarity between the expressions elicited and the ones gener- ated by algorithms. It turns out that the success of the IA depends sub- stantially on the "preference order" (po) employed by the ia, particularly in complex domains. While some pos cause the IA to produce referring expressions that are very similar to expressions produced by human sub- jects, others cause the IA to perform worse than its main competitors; moreover, it turns out to be di cult to predict the success of a po on the basis of existing psycholinguistic ndings or frequencies in corpora. We also examine the computational complexity of the algorithms in ques- tion and argue that there are no compelling reasons for preferring the ia over some of its main competitors on these grounds. We conclude that future research on the generation of referring expressions should explore alternatives to the ia, focussing on algorithms, inspired by the Greedy Algorithm, which do not work with a fixed PO.