Two Approaches for Generating Size Modifiers
Margaret A. Mitchell, Kees van Deemter, Ehud Reiter · 2011
This paper offers a solution to a small prob-lem within a much larger problem. We focus on modelling how people use size in reference, words like “big ” and “tall”, which is one piece within the much larger problem of how people refer to visible objects. Examining size in iso-lation allows us to begin untangling a few of the complex and interacting features that af-fect reference, and we isolate a set of features that may be used in a hand-coded algorithm or a machine learning approach to generate one of six basic size types. The hand-coded al-gorithm generates a modifier type with a high correspondence to those observed in human data, and achieves 81.3 % accuracy in an en-tirely new domain. This trails oracle accuracy for this task by just 8%. Features used by the hand-coded algorithm are added to a larger set of features in the machine learning approach, and we do not find a statistically significant difference between the precision and recall of the two systems. The input and output of these systems are a novel characterization of the factors that affect referring expression gen-eration, and the methods described here may serve as one building block in future work connecting vision to language. 1