Using A Text Model For Analysis And Generation

The following paper concerns a general scheme for multilingual text generation, as opposed to just translation. Our system processes the text as a whole, from which it extracts a representation of the meaning of the text. From this representation, a new text is generated, using a text modei and action ruies. This process is done in six steps : word analysis, sentence analysis using a Functional Grammar, reference soiving and inference, construction of the text pattern, sentence generation, and word generation. Different kinds of information are used st each step of the process : text organization, syntax, semantic, etc. All the knowledge, as well as the text, is given in a declarative manner. It is expressed in a single formalism named Functional Oescriptions. It consists of lexical data, a Functional Grammar, a knowledge network, action rules for reference solving and sentence generation, models of text, rules of structuration, and sentence schema. Text representation, included in the semantic network, is composed of different kinds of objects (not necessarily distinct) : text organization, syntactical information, objects introduced by the discourse, affirmations on these objects, and links between these affirmations. I TEXT VERSUS SENTENCES the one hand rearrange a whole sentence in order to respect certain stylistic rules and on the other hand modify the order of these sentences. Some of them may be deleted and others added. In any ease, in order to be able to generate a coherent text by means of the information extracted from a source text, it seems most important to us to understand what the text is refering to. We have restricted our scope to economic geography texts, taken from a French review named ATLASECO. These texts deaI with industry and agriculture, invoiving reiated concepts (growth, deciine, production, economicaI balance...). Our system parses a french text, integrates its informative content into a knowledge base, then generates a new text from information extracted from this base. We present in part (2) the text representation used throughout the process. We describe in part (3) the knowledge representation, based on the functional descriptions. In part (4), we briefly describe the different steps of the process. A text conveys information by various means. For example, the order of sentences, as well as their syntactic structure, may be significant. The general text representation we use is the same for understanding and structuring. This representation includes the different kinds of information needed for …