REFSQ 2023 hosts its renowned Doctoral Symposium (DS) for PhD students whose research relates to the field of Requirements Engineering (RE). The aimed goals for the DS are: • to provide PhD students with an opportunity to learn about the field and to get a feeling of what their colleagues are working on; • to provide PhD students with a supportive and safe environment where to present their plans and results; • to provide participating students with feedback from a panel of senior researchers in RE; • to facilitate interaction between students and established researchers in RE.
We received seven submissions; each was peer-reviewed by two members of an international panel of experts. The reviewing process focused mostly on relevance and on suitability for the doctoral symposium: not too early, not too late in the PhD trajectory. The review process led to six accepted contributions.
A slot of 45 minutes was allocated to each student: 15 minutes for presentation followed by 30 minutes of an in-depth discussion with two members of the panel. The day closed with an “ask-us-anything" panel session, in which the students could ask the panel members and Doctoral Symposium chairs for tips and advice about a PhD. Thus, the day was organized into 4 sessions. A summary of each paper, listed according to the order of presentation during the event, is ofered next. • Exploring Challenges and Solutions for Non-Functional Requirements for Machine Learning Systems, authored by Khan Mohammad Habibullah, aims at exploring quality requirements for ML systems and creating a framework for defining and managing such requirements in ML system development. • Evaluation of Quality Requirements for Explanations in AI-based Healthcare Systems, authored by Zubaria Inayat, aims at proposing a quality assessment approach to evaluate the quality of explanation for AI healthcare systems context, hence addressing explainable AI (XAI).
Session 2: User requirements • Relating User Feedback and Existing Requirements, authored by Michael Anders, aims at structuring and automatically classifying users’ feedback expressed in natural language (from fora, social media platforms, or app stores) to then use NLP techniques to link the comments with requirements. • Improving the Completeness of Acceptance Criteria, by Astrid Rohmann, follows a design science approach to develop a framework for improving the completeness of acceptance criteria for user stories through the automated recommendation of acceptance criteria. Session 3: Beyond written requirements • Conversational Requirements Engineering: Pinpointing Requirements-Relevant Information in Conversations, authored by Tjerk Spijkman, discusses the challenges and prototype solutions developed for transcribing and analyzing conversational data between practitioners captured during the initial activities of requirements engineering. • Ensuring Software Quality through Videos in Requirements Engineering, authored by Jianwei Shi, aims at reducing the gap between stakeholders and engineers, by using vision and test videos to facilitate requirements elicitation and understanding, and requirements testing.
Session 4: Ask-us-anything session In this session, the students could make questions about conducting a PhD, job perspectives after a PhD, and other PhD-related matters to some of the panel members and to the DS chairs.
For this particular Doctoral Symposium edition, we would like to thank the authors for their submissions and enthusiastic presentations and discussions during the event. Furthermore, we would like to express our gratitude to our panel of experts for their timely and quality reviews, as well as their work in guiding students during the event, namely Jean-Michel Bruel, Maya Daneva, Martin Glinz, Paul Grünbacher, Jennifer Horkof, Daniel Mendez, and Barbara Paech. Last but not least, we would like to show our appreciation to the REFSQ 2023 organizing team for their support during the whole organization process of the event.