SentiCR: A customized sentiment analysis tool for code review interactions

Ahmed, Toufique; Bosu, Amiangshu; Iqbal, Anindya; Rahimi, Shahram

doi:10.1109/ase.2017.8115623

Cited by 114 publications

(105 citation statements)

References 34 publications

Supporting

Mentioning

105

Contrasting

Order By: Relevance

“…Among the identified difficulties, lacking domain-specific knowledge is demonstrated to be the most dominant, accounting for about 81% of the classification errors. Since then, how to effectively leverage SE-related texts to introduce technical jargon becomes the main direction of the sentiment analysis in SE [8,14,41]. Against such a background, many SE-customized sentiment analysis tools and methods are proposed, including SentiStrength-SE [35], Senti-CR [8], Senti4SD [14], etc.…”

Section: Sentiment Analysis In Sementioning

confidence: 99%

“…Since then, how to effectively leverage SE-related texts to introduce technical jargon becomes the main direction of the sentiment analysis in SE [8,14,41]. Against such a background, many SE-customized sentiment analysis tools and methods are proposed, including SentiStrength-SE [35], Senti-CR [8], Senti4SD [14], etc. We take them along with the most popular SentiStrength as baseline methods in this study and introduce them detailedly in Section 4.1.…”

Section: Sentiment Analysis In Sementioning

confidence: 99%

“…As the JIRA dataset is labeled with various emotion labels, we need to map the multi-class emotions into the trinary polarities and some samples are filtered due to ambiguity. With regard to the Stack Overflow and Code Review datasets, we find both of them are manually pre-processed and filtered to follow a not that skewed sentiment distribution during their creation process [8,14]. The sampling of these datasets makes the classification tasks considerably easier, so SentiCR and SEnti-Moji do not show such obvious performance difference on these datasets.…”

Section: Rq1: How Does Sentimoji Perform Compared To the Existing Senmentioning

confidence: 99%

“…Previous studies found that sentiment analysis tools trained on non-technical texts are not adequate for SE tasks [35,38] and a lack of domain-specific knowledge is the main reason [35]. Since then, many studies focused on how to use labeled SE-related texts to train SE-customized sentiment classifiers [8,14,35]. However, in fact, there are many general sentimental expressions shared by SE and open domains.…”

Section: Lessons Learned and Implicationsmentioning

confidence: 99%

“…Furthermore, Islam and Zibran [35] applied SentiStrength to an SErelated dataset and found that misunderstanding of domain-specific meanings of words (namely technical jargon in the rest of this paper) accounts for the most misclassifications. Such a finding inspires a series of research efforts in recent years to generate SE-specific datasets and develop customized sentiment analysis methods based on them [8,14,41]. Various machine-learning or deep-learning arXiv:1907.02202v1 [cs.SE] 4 Jul 2019 techniques have been applied, but the performance still falls short of producing adequate results in practice for some SE tasks [41].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

SEntiMoji: an emoji-powered learning approach for sentiment analysis in software engineering

Chen

Cao

Lü

et al. 2019

Proceedings of the 2019 27th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of

View full text Add to dashboard Cite

Sentiment analysis has various application scenarios in software engineering (SE), such as detecting developers' emotions in commit messages and identifying their opinions on Q&A forums. However, commonly used out-of-the-box sentiment analysis tools cannot obtain reliable results on SE tasks and the misunderstanding of technical jargon is demonstrated to be the main reason. Then, researchers have to utilize labeled SE-related texts to customize sentiment analysis for SE tasks via a variety of algorithms. However, the scarce labeled data can cover only very limited expressions and thus cannot guarantee the analysis quality. To address such a problem, we turn to the easily available emoji usage data for help. More specifically, we employ emotional emojis as noisy labels of sentiments and propose a representation learning approach that uses both Tweets and GitHub posts containing emojis to learn sentiment-aware representations for SE-related texts. These emoji-labeled posts can not only supply the technical jargon, but also incorporate more general sentiment patterns shared across domains. They as well as labeled data are used to learn the final sentiment classifier. Compared to the existing sentiment analysis methods used in SE, the proposed approach can achieve significant improvement on representative benchmark datasets. By further contrast experiments, we find that the Tweets make a key contribution to the power of our approach. This finding informs future research not to unilaterally pursue the domain-specific resource, but try to transform knowledge from the open domain through ubiquitous signals such as emojis.

show abstract