MarValous - machine learning based detection of emotions in the valence-arousal space in software engineering text.
SAC(2019)
摘要
Emotion analysis in text has drawn recent interests in the software engineering (SE) community. Existing domain-independent techniques for automated emotion/sentiment analysis perform poorly when operated on SE text. Thus, a few SE domain-specific tools are recently developed for detecting sentimental polarities (e.g., positivity, negativity). But, for capturing individual emotional states such as excitation, stress, depression, and relaxation, there is only one recent tool named DEVA, which uses a lexicon-based approach.
We have developed MarValous, the first Machine Learning based tool for improved detection of the aforementioned emotional states in software engineering text. We evaluate MarValous using a dataset containing 5,122 comments collected from JIRA and Stack Overflow. From a quantitative evaluation, MarValous is found to have substantially outperformed DEVA achieving more than 83% precision and more than 79% recall.
更多查看译文
关键词
arousal, emotion, machine learning, sentiment, valence
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络