Conceptual modelling for invoice document processing

This paper is concerned with the presentation of a declarative knowledge base, the Conceptual Model, which describes the invoice domain as generally as possible. Such a model is based on a semantic network that is able to describe the invoice domain by different levels of abstraction. The Conceptual Model can be used for the labelling procedure of physical rectangles, extracted from invoices, in order to construct a model (Document Model) for each class of invoices. The Document Model contains physical coordinates for each rectangle, which can be estimated from an invoice, and the related semantic label. Once the Document Model is constructed, it can be applied to understand an invoice instance, whose class is univocally identified by its logo.