ľ¹Ï¸£ÀûÓ°ÊÓ

PLEASE NOTE: If a candidate gives a layman's talk, the livestream will start fifteen minutes earlier.

When developing software products, it is important to know exactly what users want. User feedback – such as online reviews and usage data – is a novel information source, for which this PhD dissertation proposes Crowd-based Requirements Engineering (CrowdRE). This umbrella term covers all (semi)automated approaches to gather requirements from a large group of users – a crowd. This way, software development departments can improve their products.

The PhD dissertation assesses how CrowdRE fits in with existing organizational processes, and what the impact of privacy regulations is. Most of all, the research offers solutions to analyze user feedback regarding how users feel about a software’s qualities, such as its usability, reliability, and performance. This clearly shows that users do not just want more features; they also express their opinion about qualities, whether positive or negative.

In our investigation, crowdsourcing and large language models turned out to be the most suitable methods for identifying requirements-relevant information in online user feedback. By now, CrowdRE is generally accepted in the Requirements Engineering community. As a result, the crowd has become an important source of information, while the research that is inspired by CrowdRE helps to make sure that more and better techniques for analyzing the data are becoming available. This is good news for users, because it means that their expressions on the Internet are increasingly contributing to improving the software products they use to better meet their needs.