Automatic generation of textual descriptions in data-to-text systems using a fuzzy temporal ontology: Application in air quality index data series

In this paper we present a model based on computational intelligence and natural language generation for the automatic of textual summaries from numerical data series, aiming to provide insights which help users understand relevant information hidden in data. Our includes fuzzy temporal ontology with references addresses problem managing imprecise knowledge, is series. We fully describe real use case application environmental systems field, providing linguistic descriptions about air quality index (AQI), very well-known indicator provided by all meteorological agencies worldwide. consider two different sources AQI official Galician (NW Spain) Meteorology Agency: (i) distribution stations observation network (ii) time series state evolution each station. Both models were evaluated following current standards good practices manual human expert evaluation Natural Language Generation field. Assessment results experts meteorologists satisfactory, empirically confirm that proposed fit type service both content layout.