Synthesizing illustrated documents : a plan-based approach

Abstract

The aim of our work is to develop a system able to generate documents in which text and pictures are smoothly integrated. Such tailoring requires knowledge concerning the functions of textual and pictorial document parts and the relations between them. We start from the assumption that not only the generation of text, but also the generation of multimodal documents can be considered as a sequence of communicative acts which aim to achieve certain goals. Based on textlinguistic work, the structure of an illustrated document is described by the hierarchical order of communicative acts and the relations between them. In view of the generation of text-picture combinations, we have examined relations which frequently occur between text passages and pictures, or between the parts of a picture. For the automated generation of illustrated documents, we propose a plan-based approach. To represent knowledge about presentation techniques, we have designed presentation strategies which relate to both text and picture production. Finally, we show by example how a document fragment is synthesized.