A Simple Domain-Independent Probabilistic Approach to Generation

We present a simple, robust generation system which performs content selection and surface realization in a unified, domain-independent framework. In our approach, we break up the end-to-end generation process into a sequence of local decisions, arranged hierarchically and each trained discriminatively. We deployed our system in three different domains--Robocup sportscasting, technical weather forecasts, and common weather forecasts, obtaining results comparable to state-ofthe-art domain-specific systems both in terms of BLEU scores and human evaluation.

• Type: Journal
• Year: 2010
• Where: EMNLP
• Authors: Gabor Angeli, Percy Liang, Dan Klein
• Keywords: Common Weather Forecasts, EMNLP 2010, End-to-end Generation Process, Natural Language Processing, Weather Forecasts