We describe a graph-based generation system that participated in the TUNA attribute selection and realisation task of the REG 2008 Challenge. Using a stochastic cost function (with certain properties for free), and trying attributes from cheapest to more expensive, the system achieves overall .76 DICE and .54 MASI scores for attribute selection on the development set. For realisation, it turns out that in some cases higher attribute selection accuracy leads to larger differences between system-generated and human descriptions.
|Publisher||The Association for Computational Linguistics|
|Conference||5th International Natural Language Generation Conference, INLG 2008|
|Period||12/06/08 → 14/06/08|
|Other||12-14 June 2008|