Program

Registration: https://conf.researchr.org/attending/RE-2022/registration

The workshop is currently planned to occur on Tuesday, 16th August 2022.

The times in the following program are in Melbourne(!) time (GMT+10).

Session 1: Welcome and Keynote
Chair: Alessio Ferrari

Session 2: Papers
Chair: Alessio Ferrari, Hans-Martin Heyn

Session 3: Keynote
Chair: Hans-Martin Heyn

Session 4: Papers
Chair: Mehrdad Sabetzadeh

Keynote 1

Abstract: Natural language processing and machine/deep learning have been widely used by RE researchers for the automated analysis of written artefacts, notably requirements specifications, which play a pivotal role in RE processes. Many key RE activities, however, are rooted in synchronous conversations rather than written documents: elicitation interviews and workshops, refinement meetings, and validation sessions, to name a few. Very limited research exists that applies AI techniques to conversations. Most likely, this has to do with the fact that researchers can more easily gain access to specifications than to recorded and transcribed conversations. The availability of digital communication tools that allow for automated recording and transcription (e.g., Teams and Zoom), together with the increasing use of these tools in RE activities, offers an opportunity for RE researchers to automatically analyze requirements conversations. This is especially interesting in agile development settings, where minimal documentation is preferred to comprehensive documentation. In this talk, I will present use cases for the analysis of RE conversations, explore the unique challenges of this setting, and discuss ongoing research concerning the identification of requirements-relevant information in RE conversations.

About the speaker: Dr. Fabiano Dalpiaz is an associate professor in the Department of Information and Computing Sciences at Utrecht University in the Netherlands. He is principal investigator in the department's Requirements Engineering lab. In his research, Fabiano blends artificial intelligence with information visualization in order to increase the quality of the requirements engineering process and artifacts, with the ultimate aim of delivering higher-quality software. His research is often validated in-vivo through collaborations with the software industry. He was program co-chair of REFSQ 2021 and RCIS 2020, and will be PC co-chair of IEEE RE 2023. He was the organization chair of the REFSQ 2018 conference, and he is an associate editor for the Requirements Engineering Journal and the Business & Information Systems Engineering Journal. He often serves on the program committee of conferences such as RE, CAiSE, REFSQ, ICSE, and AAMAS.

Keynote 2

Abstract: As the use of AI and machine learning (ML) in business and research becomes more commonplace, we as the requirements engineering community must understand how our knowledge and experience can apply, else risk being left out of the growing paradigm of applied ML. As RE researchers, we must ask ourselves: how can our concepts, abstractions and methods be applied – or not? We are no longer able to completely define or accurately predict the behaviors or outcomes of the systems we use – yet we still need a way to define feature and system success or failure. As one part of this larger challenge, one can argue that requirements for ML-enabled systems are less about what the system does and more about how well the system achieves various qualities (e.g., performance, fairness, reliability). For this, we can turn to our traditional notion of quality requirements or non-functional requirements (NFRs). However, NFRs for ML-enabled systems may be different than for traditional systems: different NFRs may be important (e.g., fairness), NFR meanings and measures may be different (e.g., performance), and new and old concepts may live together in the same complex systems. In this talk, I present our recent work examining NFRs for ML, including academic emphasis via early systematic review results, and the perception of NFRs for ML from industrial and academic interview and survey respondents. I give an overview of a concrete application case -- perception systems for autonomous driving -- where we explore industrial views on requirements processes for ML-enabled systems, new and old NFRs, tradeoffs, and the link between data quality and the quality of system features. I hope that this talk fosters discussion and collaboration on RE for AI and ML.

About the speaker: Jennifer Horkoff is an Associate Professor at the Interaction Design and Software Engineering division in the Computer Science and Engineering Department shared by Chalmers University of Technology and the University of Gothenburg, Sweden. Dr. Horkoff is currently involved in projects investigating non-functional requirements for machine learning (supported by the Swedish Research Council), the role of RE and conceptual modeling in large-scale Agile (supported by the Chalmers Software Center in collaboration with local industry), and applying deep learning to understand software architecture (supported by Chalmers AI Research Centre). Jennifer received her Ph.D. in Computer Science from the University of Toronto in 2012 and received her Swedish Docent in 2020. She has been an author or co-author of more than 70 papers in peer-reviewed journals, conferences, or workshops. Jennifer has been a program co-chair of REFSQ, PoEM, and ER and will be the co-program chair of RE 2023.

---