User needs inform designers and developers of essential functionalities for requirements engineering. In this work, we summarize key concepts and challenges relating to manual and automatic user needs detection methods. We discuss six challenges with manual and eight challenges with automated
methods.
A. Challenges of Manual Methods
• A01: Limited sample sizes.
• A02: Budgetary constraints.
• A03: Time constraints.
• A04: Lack of scalability.
• A05: Difficulty of needs articulation.
• A06: Human bias.
B. Challenges of Automated Methods
• B01: Inadequate datasets.
• B02: Irrelevant information.
• B03: Lack of informativeness.
• B04: Social desirability bias.
• B05: Lack of standards.
• B06: Lack of applicability.
• B07: Inability to determine meaning.
• B08: Need for human involvement.
Despite the promise of automated methods, the challenges imply that artificial intelligence and machine learning are not yet mature enough to replace manual methods, such as interviews and focus groups, for discovering user needs in requirements engineering.
Salminen, J., Jung, S.G., and Jansen, B. J (2021) Manual and Automatic Methods for User Needs Detection in Requirements Engineering: Key Concepts and Challenges. The International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME2021). 7-8 October. Mauritius. p. 1-7.