The majority of requirements are given in natural language, either written or orally expressed. Other requirements might also be visually expressed in terms of figures, diagrams, images or even gestures. Artificial-intelligence approaches might be used to develop prototypes, which can then be re-engineered using more conventional requirements technologies and safety assurance techniques. For example, we might employ large amounts of semantic and statistical data, knowledge bases and theorem provers to infer as much contextual information as possible from the (vague) textual or visual requirements. The automatic analysis of natural language expressions has not yet been fully achieved, and interdisciplinary methodologies and tools are needed to successfully go from natural language to accurate formal specifications. Conformance of a system implementation to its requirements requires dynamic and efficient communication and iteration among system stakeholders. It is in supporting this process, and not in supplanting it, that innovative approaches to requirements analysis need to find their proper role.

We want to gain a better understanding about how to deal with natural language as the vehicle from which we derive system/software requirements, how to use intelligent agents as entities to facilitate semi-automatic requirements-documentation analysis, and how to build automatic systems to aid in requirements/specifications elicitation.

A good case study for these issues would be to consider how to extract a conceptual model of the goals and requirements of the software needs discussed in a blog.  As blogs are unstructured natural language, it represents one of the most difficult challenges for natural language processing.

In this workshop we will bring together experienced researchers who have been involved in the specification, design, development and validation/verification of software-intensive systems, both for software engineering and for natural language processing. The overall aim is to exchange ideas for continued research in the intersection of these two areas and to reduce the gap between theory and practice.  

The last few years the workshop has been held alternately in Europe and the United States. This year, the workshop will be held in Monterey, California on September 10-13, 2007.

Workshop topics include but are not limited to:

Proceedings

Given the interactive nature of the workshop, the proceedings will be published after the workshop to allow the participants to include the results of the discussion in their submissions. All submissions will be reviewed by the program committee before publication.

The workshop agenda and abstracts of presentations will be published on this website before the workshop, and presentation material will be available on this site afterwards, similar to the websites of previous workshops in this series (e.g., see for the 2006 workshop’s website as well as links to the websites of previous years, which can be reached from the 2007 edition’s web site).