Discourse structure analysis for requirement mining

Juyeon Kang (1) , Patrick Saint Dizier (2)

1 IRIT-SAMoVA - Équipe Structuration, Analyse et MOdélisation de documents Vidéo et Audio
2 IRIT-ADRIA - Argumentation, Décision, Raisonnement, Incertitude et Apprentissage

In this work, we first introduce two main approaches to writing requirements and then propose a method based on Natural Language Processing to improve requirement authoring and the overall coherence, cohesion and organization of requirement documents. We investigate the structure of requirement kernels, and then the discourse structure associated with those kernels. This will then enable the system to accurately extract requirements and their related contexts from texts (called requirement mining). Finally, we relate a first experimentation on requirement mining based on texts from seven companies. An evaluation that compares those results with manually annotated corpora of documents is given to conclude.

Mots clés: Requirement Authoring; Requirement Mining; Discourse Structure Analysis; Natural Language Processing

Domaines: Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]; Logique en informatique [cs.LO]; Informatique et langage [cs.CL]

Juyeon Kang, Patrick Saint Dizier. Discourse structure analysis for requirement mining. International Journal of Knowledge Content Development & Technology, 2013, vol. 3 (n° 2), pp.43-65. ⟨10.5865/IJKCT.2013.3.2.043⟩. ⟨hal-01137321⟩