Linguistic Analysis of Requirements of a Space Project and Their Conformity with the Recommendations Proposed by a Controlled Natural Language

Résumé En

We propose a linguistic analysis of requirements written in French for a project carried out by the French National Space Agency (CNES). The aim is to determine to what extent they conform to some of the rules laid down in INCOSE, a recent guide for writing requirements, with a focus on the notion of sentence " comprehensibility ". Although CNES engineers are not obliged to follow any Controlled Natural Language, we believe that language regularities are likely to emerge from this task, mainly due to the writers' experience. As a first step, we use natural language processing tools to identify sentences that do not comply with INCOSE rules. We further review these sentences to understand why the recommendations cannot (or should not) always be applied when specifying large-scale projects, and how they could be improved. This paper presents a corpus linguistics approach applied to the melioration of requirements writing. We propose a linguistic diagnosis of the way requirements are written in a space project by comparing these requirements with a guide for writing specifications (a controlled natural language). Initial results obtained from this analysis suggest that guides for writing specifications are not fully adapted to the real writing process: they are sometimes too constraining, and sometimes insufficiently so. In the medium term, the aim is to propose another guide based on the spontaneous regularities observed in requirements. The paper comprises two parts. In the first one (see section 2), we present the context of our study and the tool-assisted method used for making the diagnosis. In the second one (see section 3), we describe and discuss our preliminary results.