Search

In this paper, we present a novel self-supervised framework for Sentiment Cue Extraction (SCE) aimed at enhancing the interpretability of text sentiment analysis models. Our approach leverages self-supervised learning to identify and highlight key textual elements that significantly influence sentiment classification decisions. Central to our framework is the development of an innovative Mask Sequence Interpretation Score (MSIS), a bespoke metric designed to assess the relevance and coherence of identified sentiment cues within binary text classification tasks. By employing Monte Carlo Sampling techniques optimized for computational efficiency, our framework demonstrates exceptional effectiveness in processing large-scale text data across diverse datasets, including English and Chinese, thus proving its versatility and scalability. The effectiveness of our approach is validated through extensive experiments on several benchmark datasets, including SST-2, IMDb, Yelp, and ChnSentiCorp. The results indicate a substantial improvement in the interpretability of the sentiment analysis models without compromising their predictive accuracy. Furthermore, our method stands out for its global interpretability, offering an efficient solution for analyzing new data compared to traditional techniques focused on local explanations.

Emotion analysis is currently a popular research direction in the field of natural language processing. However, existing research focuses primarily on tasks such as emotion classification, emotion extraction, and emotion cause analysis, while there are few investigations into the relationship between emotions and their impacts. To address these limitations, this paper introduces the emotion-driven behavior extraction (EDBE) task, which addresses these limitations by separately extracting emotions and behaviors to filter emotion-driven behaviors described in text. EDBE comprises three sub-tasks: emotion extraction, behavior extraction, and emotion–behavior pair filtering. To facilitate research in this domain, we have created a new dataset, which is accessible to the research community. To address the EDBE task, we propose a pipeline approach that incorporates the causal relationship between emotions and driven behaviors. Additionally, we adopt the prompt paradigm to improve the model’s representation of cause-and-effect relationships. In comparison to state-of-the-art methods, our approach demonstrates notable improvements, achieving a 1.32% improvement at the clause level and a 1.55% improvement at the span level on our newly curated dataset in terms of the F1 score, which is a commonly used metric to measure the performance of models. These results underscore the effectiveness and superiority of our approach in relation to existing methods.

BackgroundCOVID-19 vaccines are one of the most effective preventive strategies for containing the pandemic. Having a better understanding of the public’s conceptions of COVID-19 vaccines may aid in the effort to promptly and thoroughly vaccinate the community. However, because no empirical research has yet fully explored the public’s vaccine awareness through sentiment–based topic modeling, little is known about the evolution of public attitude since the rollout of COVID-19 vaccines. ObjectiveIn this study, we specifically focused on tweets about COVID-19 vaccines (Pfizer, Moderna, AstraZeneca, and Johnson & Johnson) after vaccines became publicly available. We aimed to explore the overall sentiments and topics of tweets about COVID-19 vaccines, as well as how such sentiments and main concerns evolved. MethodsWe collected 1,122,139 tweets related to COVID-19 vaccines from December 14, 2020, to April 30, 2021, using Twitter’s application programming interface. We removed retweets and duplicate tweets to avoid data redundancy, which resulted in 857,128 tweets. We then applied sentiment–based topic modeling by using the compound score to determine sentiment polarity and the coherence score to determine the optimal topic number for different sentiment polarity categories. Finally, we calculated the topic distribution to illustrate the topic evolution of main concerns. ResultsOverall, 398,661 (46.51%) were positive, 204,084 (23.81%) were negative, 245,976 (28.70%) were neutral, 6899 (0.80%) were highly positive, and 1508 (0.18%) were highly negative sentiments. The main topics of positive and highly positive tweets were planning for getting vaccination (251,979/405,560, 62.13%), getting vaccination (76,029/405,560, 18.75%), and vaccine information and knowledge (21,127/405,560, 5.21%). The main concerns in negative and highly negative tweets were vaccine hesitancy (115,206/205,592, 56.04%), extreme side effects of the vaccines (19,690/205,592, 9.58%), and vaccine supply and rollout (17,154/205,592, 8.34%). During the study period, negative sentiment trends were stable, while positive sentiments could be easily influenced. Topic heatmap visualization demonstrated how main concerns changed during the current widespread vaccination campaign. ConclusionsTo the best of our knowledge, this is the first study to evaluate public COVID-19 vaccine awareness and awareness trends on social media with automated sentiment–based topic modeling after vaccine rollout. Our results can help policymakers and research communities track public attitudes toward COVID-19 vaccines and help them make decisions to promote the vaccination campaign.

Emotion entrainment, which is generally defined as the synchronous convergence of human emotions, performs many important social functions. However, what the specific mechanisms of emotion entrainment are beyond in-person interactions, and how human emotions evolve under different entrainment patterns in large-scale social communities, are still unknown. In this paper, we aim to examine the massive emotion entrainment patterns and understand the underlying mechanisms in the context of social media. As modeling emotion dynamics on a large scale is often challenging, we elaborate a pragmatic framework to characterize and quantify the entrainment phenomenon. By applying this framework on the datasets from two large-scale social media platforms, we find that the emotions of online users entrain through social networks. We further uncover that online users often form their relations via dual entrainment, while maintain it through single entrainment. Remarkably, the emotions of online users are more convergent in nonreciprocal entrainment. Building on these findings, we develop an entrainment augmented model for emotion prediction. Experimental results suggest that entrainment patterns inform emotion proximity in dyads, and encoding their associations promotes emotion prediction. This work can further help us to understand the underlying dynamic process of large-scale online interactions and make more reasonable decisions regarding emergency situations, epidemic diseases, and political campaigns in cyberspace.