2021/08/04 by Marc Herrmann, Herrmann, Marc, Jil Klünder +1 · 2 citations
Computer Science · #FOS: Computer and information sciences #Software Engineering (cs.SE) #Software Engineering Research #Software Engineering Techniques and Practices #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2108.01985
openalex publication_date 2021/08/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Sentiment analysis gets increasing attention in software engineering with new\ntools emerging from new insights provided by researchers. Existing use cases\nand tools are meant to be used for textual communication such as comments on\ncollaborative version control systems. While this can already provide useful\nfeedback for development teams, a lot of communication takes place in meetings\nand is not suited for present tool designs and concepts.\n In this paper, we present a concept that is capable of processing live\nmeeting audio and classifying transcribed statements into sentiment polarity\nclasses. We combine the latest advances in open source speech recognition with\nprevious research in sentiment analysis. We tested our approach on a student\nsoftware project meeting to gain proof of concept, showing moderate agreement\nbetween the classifications of our tool and a human observer on the meeting\naudio. Despite the preliminary character of our study, we see promising results\nmotivating future research in sentiment analysis on meetings. For example, the\npolarity classification can be extended to detect destructive behaviour that\ncan endanger project success.\n