What is it about?
Bots are software systems that automate tasks for their users. When they talk with users in natural language, they are also called conversational agents or chatbots. With generative AI and large language models, bots are used more and more in software development. Still, practitioners report that bots can sometimes add difficulties instead of helping. In this work, we carried out a multivocal literature review: we reviewed both research papers and practitioner sources. We used them to describe how bots can be characterized and what challenges teams face when adopting them in software engineering.
Featured Image
Photo by Ant Rozetsky on Unsplash
Why is it important?
We provide a taxonomy to characterize bots, a list of challenges for their adoption in software engineering, and strategies to mitigate them. Researchers can use our results to choose future research directions, and teams can use the strategies to improve how they use bots. The review also helps move knowledge from research to practice.
Perspectives
Bots and conversational agents are now common in software development, even more with generative AI, but practitioners also report that they can make work harder. In this review, we looked at both research papers and practitioner sources, because many lessons about bots come from industry. I hope the taxonomy and the list of challenges and strategies help both researchers and teams that want to adopt bots.
Prof. Stefano Lambiase
Aalborg Universitet
Read the Original
This page is a summary of: Motivations, Challenges, Best Practices, and Benefits for Bots and Conversational Agents in Software Engineering: A Multivocal Literature Review, ACM Computing Surveys, November 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3704806.
You can read the full text:
Contributors
The following have contributed to this page
