Probabilistic Generation of Weather Forecast Texts

This paper reports experiments in which pCRU — a generation framework that combines probabilistic generation methodology with a comprehensive model of the generation space — is used to semi-automatically create several versions of a weather forecast text generator. The generators are evaluated in terms of output quality, development time and computational efficiency against (i) human forecasters, (ii) a traditional handcrafted pipelined NLG system, and (iii) a HALOGEN-style statistical generator. The most striking result is that despite acquiring all decision-making abilities automatically, the best pCRU generators receive higher scores from human judges than forecasts written by experts.