CNTS: Memory-Based learning of generating repeated references
Iris Hendrickx, Walter Daelemans, Kim Luyckx, Roser Morante and Vincent Van Asch
In: 5th Natural Language Generation Conference, 12-14 June 2008, Salt Fork, Ohio, USA.

Abstract

In this paper we describe our machine learning approach to the generation of referring expressions. As our algorithm we use memory-based learning. Our results show that in case of predicting the T YP E of the expression, having one general classifier gives the best results. On the contrary, when predicting the full set of properties of an expression, a combined set of specialized classifiers for each subdomain gives the best performance. ing as implemented in the Timbl package (Daelemans et al., 2007). To select the optimal algorith- mic parameter setting for each classifier we used a heuristic optimization method called paramsearch (Van den Bosch, 2004). We also tried several other machine learning algorithms implemented in the Weka package (Witten and Frank, 2005), but these experiments did not lead to better results and are not

[Edit]