Surface realization architecture for low-resourced African languages

Abstract

There has been growing interest in building surface realization systems to support the automatic generation of text in African languages. Such tools focus on converting abstract representations of meaning to text. Since African languages are low-resourced, economical use of resources and general maintainability are key considerations. However, there is no existing surface realizer architecture that possesses most of the maintainability characteristics (e.g., modularity, reusability, and analyzability) that will lead to maintainable software that can be used for the languages. Moreover, there is no consensus surface realization architecture created for other languages that can be adapted for the languages in question. In this work, we solve this by creating a novel surface realizer architecture suitable for low-resourced African languages that abide by the features of maintainable software. Its design comes after a granular analysis, classification, and comparison of the architectures used by 77 existing NLG systems. We compare our architecture to existing architectures and show that it supports the most features of a maintainable software product.