RESUMEN

Zero-shot stance detection is pivotal for autonomously discerning user stances on novel emerging topics. This task hinges on effective feature alignment transfer from known to unseen targets. To address this, we introduce a zero-shot stance detection framework utilizing multi-expert cooperative learning. This framework comprises two core components: a multi-expert feature extraction module and a gating mechanism for stance feature selection. Our approach involves a unique learning strategy tailored to decompose complex semantic features. This strategy harnesses the expertise of multiple specialists to unravel and learn diverse, intrinsic textual features, enhancing transferability. Furthermore, we employ a gating-based mechanism to selectively filter and fuse these intricate features, optimizing them for stance classification. Extensive experiments on standard benchmark datasets demonstrate that our model significantly surpasses existing baseline models in performance.

RESUMEN

During the COVID-19 pandemic, nations implemented various preventive measures, triggering varying online responses. This study examines cultural influences on public online stances toward these measures and their impacts on COVID-19 cases/deaths. Stance detection analysis was used to analyze 16,428,557 Tweets regarding COVID-19 preventive measures from 95 countries, selected based on Hofstede's cultural dimensions. To ensure the variety of population, countries were chosen based on Twitter data availability and a minimum sample size of 385 tweets, achieving a 95% confidence level with a 5% margin of error. The weighted regression analysis revealed that the relationship between culture and online stances depends on the cultural congruence of each measure. Specifically, power distance positively predicted stances for all measures, while indulgence had a negative effect overall. Effects of other cultural indices varied across measures. Individualism negatively affected face coverings stances. Uncertainty avoidance influenced lockdown and vaccination stances negatively but had a positive effect on social distancing stances. Long-term orientation negatively affected lockdown and social distancing stances but positively influenced quarantine stances. Cultural tightness only negatively affected face coverings and quarantine stances. Online stances toward face coverings mediated the relationship between cultural indices and COVID-19 cases/deaths. As such, public health officials should consider cultural profiles and use culturally congruent communication strategies when implementing preventive measures for future pandemics. Furthermore, leveraging digital tools is vital in navigating and shaping online stances to enhance the effectiveness of these measures.

RESUMEN

In social networks, the occurrence of unexpected events rapidly catalyzes the widespread dissemination and further evolution of network public opinion. The advent of zero-shot stance detection aligns more closely with the characteristics of stance detection in today's digital age, where the absence of training examples for specific models poses significant challenges. This task necessitates models with robust generalization abilities to discern target-related, transferable stance features within training data. Recent advances in prompt-based learning have showcased notable efficacy in few-shot text classification. Such methods typically employ a uniform prompt pattern across all instances, yet they overlook the intricate relationship between prompts and instances, thereby failing to sufficiently direct the model towards learning task-relevant knowledge and information. This paper argues for the critical need to dynamically enhance the relevance between specific instances and prompts. Thus, we introduce a stance detection model underpinned by a gated multilayer perceptron (gMLP) and a prompt learning strategy, which is tailored for zero-shot stance detection scenarios. Specifically, the gMLP is utilized to capture semantic features of instances, coupled with a control gate mechanism to modulate the influence of the gate on prompt tokens based on the semantic context of each instance, thereby dynamically reinforcing the instance-prompt connection. Moreover, we integrate contrastive learning to empower the model with more discriminative feature representations. Experimental evaluations on the VAST and SEM16 benchmark datasets substantiate our method's effectiveness, yielding a 1.3% improvement over the JointCL model on the VAST dataset.

RESUMEN

The COVID-19 pandemic has sparked numerous discussions on social media platforms, with users sharing their views on topics such as mask-wearing and vaccination. To facilitate the evaluation of neural models for stance detection and premise classification, we organized the Social Media Mining for Health (SMM4H) 2022 Shared Task 2. This competition utilized manually annotated posts on three COVID-19-related topics: school closures, stay-at-home orders, and wearing masks. In this paper, we extend the previous work and present newly collected data on vaccination from Twitter to assess the performance of models on a different topic. To enhance the accuracy and effectiveness of our evaluation, we employed various strategies to aggregate tweet texts with claims, including models with feature-level (early) fusion and dual-view architectures from the SMM4H 2022 Task 2 leaderboard. Our primary objective was to create a valuable dataset and perform an extensive experimental evaluation to support future research in argument mining in the health domain.

Asunto(s)

RESUMEN

BACKGROUND: Developing an understanding of the public discourse on COVID-19 vaccination on social media is important not only for addressing the ongoing COVID-19 pandemic but also for future pathogen outbreaks. There are various research efforts in this domain, although, a need still exists for a comprehensive topic-wise analysis of tweets in favor of and against COVID-19 vaccines. OBJECTIVE: This study characterizes the discussion points in favor of and against COVID-19 vaccines posted on Twitter during the first year of the pandemic. The aim of this study was primarily to contrast the views expressed by both camps, their respective activity patterns, and their correlation with vaccine-related events. A further aim was to gauge the genuineness of the concerns expressed in antivax tweets. METHODS: We examined a Twitter data set containing 75 million English tweets discussing the COVID-19 vaccination from March 2020 to March 2021. We trained a stance detection algorithm using natural language processing techniques to classify tweets as antivax or provax and examined the main topics of discourse using topic modeling techniques. RESULTS: Provax tweets (37 million) far outnumbered antivax tweets (10 million) and focused mostly on vaccine development, whereas antivax tweets covered a wide range of topics, including opposition to vaccine mandate and concerns about safety. Although some antivax tweets included genuine concerns, there was a large amount of falsehood. Both stances discussed many of the same topics from opposite viewpoints. Memes and jokes were among the most retweeted messages. Most tweets from both stances (9,007,481/10,566,679, 85.24% antivax and 24,463,708/37,044,507, 66.03% provax tweets) came from dual-stance users who posted both provax and antivax tweets during the observation period. CONCLUSIONS: This study is a comprehensive account of COVID-19 vaccine discourse in the English language on Twitter from March 2020 to March 2021. The broad range of discussion points covered almost the entire conversation, and their temporal dynamics revealed a significant correlation with COVID-19 vaccine-related events. We did not find any evidence of polarization and prevalence of antivax discourse over Twitter. However, targeted countering of falsehoods is important because only a small fraction of antivax discourse touched on a genuine issue. Future research should examine the role of memes and humor in driving web-based social media activity.

Asunto(s)

RESUMEN

BACKGROUND: Since the onset of the COVID-19 pandemic, vaccines have been an important topic in public discourse. The discussions around vaccines are polarized, as some see them as an important measure to end the pandemic, and others are hesitant or find them harmful. A substantial portion of these discussions occurs openly on social media platforms. This allows us to closely monitor the opinions of different groups and their changes over time. OBJECTIVE: This study investigated posts related to COVID-19 vaccines on Twitter (Twitter Inc) and focused on those that had a negative stance toward vaccines. It examined the evolution of the percentage of negative tweets over time. It also examined the different topics discussed in these tweets to understand the concerns and discussion points of those holding a negative stance toward the vaccines. METHODS: A data set of 16,713,238 English tweets related to COVID-19 vaccines was collected, covering the period from March 1, 2020, to July 31, 2021. We used the scikit-learn Python library to apply a support vector machine classifier to identify the tweets with a negative stance toward COVID-19 vaccines. A total of 5163 tweets were used to train the classifier, of which a subset of 2484 tweets was manually annotated by us and made publicly available along with this paper. We used the BERTopic model to extract the topics discussed within the negative tweets and investigate them, including how they changed over time. RESULTS: We showed that the negativity with respect to COVID-19 vaccines has decreased over time along with the vaccine rollouts. We identified 37 topics of discussion and presented their respective importance over time. We showed that popular topics not only consisted of conspiratorial discussions, such as 5G towers and microchips, but also contained legitimate concerns around vaccination safety and side effects as well as concerns about policies. The most prevalent topic among vaccine-hesitant tweets was related to the use of messenger RNA and fears about its speculated negative effects on our DNA. CONCLUSIONS: Hesitancy toward vaccines existed before the COVID-19 pandemic. However, given the dimension of and circumstances surrounding the COVID-19 pandemic, some new areas of hesitancy and negativity toward COVID-19 vaccines have arisen, for example, whether there has been enough time for them to be properly tested. There is also an unprecedented number of conspiracy theories associated with them. Our study shows that even unpopular opinions or conspiracy theories can become widespread when paired with a widely popular discussion topic such as COVID-19 vaccines. Understanding the concerns, the discussed topics, and how they change over time is essential for policy makers and public health authorities to provide better in-time information and policies to facilitate the vaccination of the population in future similar crises.

Asunto(s)

RESUMEN

Stance detection is an evolving opinion mining research area motivated by the vast increase in the variety and volume of user-generated content. In this regard, considerable research has been recently carried out in the area of stance detection. In this study, we review the different techniques proposed in the literature for stance detection as well as other applications such as rumor veracity detection. Particularly, we conducted a systematic literature review of empirical research on the machine learning (ML) models for stance detection that were published from January 2015 to October 2022. We analyzed 96 primary studies, which spanned eight categories of ML techniques. In this paper, we categorize the analyzed studies according to a taxonomy of six dimensions: approaches, target dependency, applications, modeling, language, and resources. We further classify and analyze the corresponding techniques from each dimension's perspective and highlight their strengths and weaknesses. The analysis reveals that deep learning models that adopt a mechanism of self-attention have been used more frequently than the other approaches. It is worth noting that emerging ML techniques such as few-shot learning and multitask learning have been used extensively for stance detection. A major conclusion of our analysis is that despite that ML models have shown to be promising in this field, the application of these models in the real world is still limited. Our analysis lists challenges and gaps to be addressed in future research. Furthermore, the taxonomy presented can assist researchers in developing and positioning new techniques for stance detection-related applications.

RESUMEN

As a platform of social media with high activity, Twitter has seen the discussion of many hot topics related to the COVID-19 pandemic. One such is the COVID-19 vaccination program, which has skeptics in several religious, ethnic, and socioeconomic groups, and Indonesia has one of the largest populations of various ethnicities and religions of countries worldwide. Diverse opinions based on skepticism about the effectiveness of vaccines can increase the number of people who refuse or delay vaccine acceptance. Therefore, it is important to analyze and monitor stances and public opinions on social media, especially on vaccine topics, as part of the long-term solution to the COVID-19 pandemic. This study presents the Indonesian COVID-19 vaccine-related tweets data set that contains stance and aspect-based sentiment information. The data were collected monthly from January to October 2021 using specific keywords. There are nine thousand tweets manually annotated by three independent analysts. We annotated each tweet with three labels of stance and seven predetermined aspects related to Indonesian COVID-19 vaccine-related tweets: services, implementation, apps, costs, participants, vaccine products, and general. The dataset is useful for many research purposes, including stance detection, aspect-based sentiment analysis, topic detection, and public opinion analysis on Twitter, especially on the policies regarding the prevention of pandemics.

RESUMEN

The coronavirus has caused significant disruption to people's everyday lives, altering how people live, work, and study. The Kingdom of Saudi Arabia (KSA) reacted very quickly to suppress the spread of the virus even before the first case of COVID-19 was confirmed in the country. In the education sector, all face-to-face activities at public and private schools and universities were suspended, as they switched from traditional to distance learning for the entire 2020 academic year. This study collected 1,846,285 tweets to analyze the public's dynamic opinions towards distance education in the KSA during the 2020 academic year. Several classical machine-learning models and deep-learning models, including ensemble random forest (RF), support vector machine (SVM), adaptive boosting (AdaBoost), multinomial naïve Bayes (MNB), convolutional neural network (CNN), and long short-term memory (LSTM), were tested on this data, and the best-performing models were selected to analyze the public stance towards distance education. Additionally, I correlated my analysis with the major events that were announced by the Ministry of Education (MOE). I observed that people in the KSA took some time to react and express their stances at the start of the academic year. Regarding the news, I observed that any exam-related topic attracted high engagement. In-favor stances increased when news headlines covered the topic of exams compared to other topics. The results show that the primary Saudi public stance favored distance education during the 2020 academic year.

Asunto(s)

RESUMEN

Facing the COVID-19 pandemic, governments have implemented a wide range of policies to contain the spread of the virus. During the pandemic, large amounts of COVID-19-related tweets emerge every day. Real-time processing of daily tweets may offer insights for monitoring public opinion about intervention measures implemented. In this work, lockdown policy in New York State has been set as a target of public opinion research. This task includes two stages, stance detection and opinion monitoring. For the stance detection stage, we explored several combinations of different text representations and classification algorithms, finding that the combination of Long Short-Term Memory (LSTM) with Global Vectors for Word Representation (GloVe) outperforms others. Due to the shortage of labeled data, we adopted the data distillation method for the training data augmentation. The augmentation of the training data allows to improve the performance of the model with a very small amount of manually-labeled data. After applying the distillation method, the accuracy of the model has been significantly improved. Utilizing the enhanced model, automatically classified tweets are analyzed over time to monitor the public opinion. By exploring the tweets in New York from January 22nd until September 30th, 2020, we show the correlation of public opinion with COVID-19 cases and mortality data, and the effect of government responses on the opinion shift. These results demonstrate the capability of the presented method to effectively and efficiently monitor public opinion during a pandemic.

RESUMEN

A major challenge in stance detection is the large (potentially infinite) and diverse set of stance topics. Collecting data for such a set is unrealistic due to both the expense of annotation and the continuous creation of new real-world topics (e.g., a new politician runs for office). Furthermore, stancetaking occurs in a wide range of languages and genres (e.g., Twitter, news articles). While zero-shot stance detection in English, where evaluation is on topics not seen during training, has received increasing attention, we argue that this attention should be expanded to multilingual and multi-genre settings. We discuss two paradigms for English zero-shot stance detection evaluation, as well as recent work in this area. We then discuss recent work on multilingual and multi-genre stance detection, which has focused primarily on non-zero-shot settings. We argue that this work should be expanded to multilingual and multi-genre zero-shot stance detection and propose best practices to systematize and stimulate future work in this direction. While domain adaptation techniques are well-suited for work in these settings, we argue that increased care should be taken to improve model explainability and to conduct robust evaluations, considering not only empirical generalization ability but also the understanding of complex language and inferences.

RESUMEN

Recent breakthroughs in machine learning and big data analysis are allowing our online activities to be scrutinized at an unprecedented scale, and our private information to be inferred without our consent or knowledge. Here, we focus on algorithms designed to infer the opinions of Twitter users toward a growing number of topics, and consider the possibility of modifying the profiles of these users in the hope of hiding their opinions from such algorithms. We ran a survey to understand the extent of this privacy threat, and found evidence suggesting that a significant proportion of Twitter users wish to avoid revealing at least some of their opinions about social, political, and religious issues. Moreover, our participants were unable to reliably identify the Twitter activities that reveal one's opinion to such algorithms. Given these findings, we consider the possibility of fighting AI with AI, i.e., instead of relying on human intuition, people may have a better chance at hiding their opinion if they modify their Twitter profiles following advice from an automated assistant. We propose a heuristic that identifies which Twitter accounts the users should follow or mention in their tweets, and show that such a heuristic can effectively hide the user's opinions. Altogether, our study highlights the risk associated with developing machine learning algorithms that analyze people's profiles, and demonstrates the potential to develop countermeasures that preserve the basic right of choosing which of our opinions to share with the world.

RESUMEN

Background: The spread of rumors related to COVID-19 on social media has posed substantial challenges to public health governance, and thus exposing rumors and curbing their spread quickly and effectively has become an urgent task. This study aimed to assist in formulating effective strategies to debunk rumors and curb their spread on social media. Methods: A total of 2,053 original postings and 100,348 comments that replied to the postings of five false rumors related to COVID-19 (dated from January 20, 2020, to June 28, 2020) belonging to three categories, authoritative, social, and political, on Sina Weibo in China were randomly selected. To study the effectiveness of different debunking methods, a new annotation scheme was proposed that divides debunking methods into six categories: denial, further fact-checking, refutation, person response, organization response, and combination methods. Text classifiers using deep learning methods were built to automatically identify four user stances in comments that replied to debunking postings: supporting, denying, querying, and commenting stances. Then, based on stance responses, a debunking effectiveness index (DEI) was developed to measure the effectiveness of different debunking methods. Results: The refutation method with cited evidence has the best debunking effect, whether used alone or in combination with other debunking methods. For the social category of Car rumor and political category of Russia rumor, using the refutation method alone can achieve the optimal debunking effect. For authoritative rumors, a combination method has the optimal debunking effect, but the most effective combination method requires avoiding the use of a combination of a debunking method where the person or organization defamed by the authoritative rumor responds personally and the refutation method. Conclusion: The findings provide relevant insights into ways to debunk rumors effectively, support crisis management of false information, and take necessary actions in response to rumors amid public health emergencies.

Asunto(s)

RESUMEN

The evolution of electronic media is a mixed blessing. Due to the easy access, low cost, and faster reach of the information, people search out and devour news from online social networks. In contrast, the increasing acceptance of social media reporting leads to the spread of fake news. This is a minacious problem that causes disputes and endangers the societal stability and harmony. Fake news spread has gained attention from researchers due to its vicious nature. proliferation of misinformation in all media, from the internet to cable news, paid advertising and local news outlets, has made it essential for people to identify the misinformation and sort through the facts. Researchers are trying to analyze the credibility of information and curtail false information on such platforms. Credibility is the believability of the piece of information at hand. Analyzing the credibility of fake news is challenging due to the intent of its creation and the polychromatic nature of the news. In this work, we propose a model for detecting fake news. Our method investigates the content of the news at the early stage i.e., when the news is published but is yet to be disseminated through social media. Our work interprets the content with automatic feature extraction and the relevance of the text pieces. In summary, we introduce stance as one of the features along with the content of the article and employ the pre-trained contextualized word embeddings BERT to obtain the state-of-art results for fake news detection. The experiment conducted on the real-world dataset indicates that our model outperforms the previous work and enables fake news detection with an accuracy of 95.32%.

RESUMEN

Reconstructing a "forma mentis", a mindset, and its changes, means capturing how individuals perceive topics, trends and experiences over time. To this aim we use forma mentis networks (FMNs), which enable direct, microscopic access to how individuals conceptually perceive knowledge and sentiment around a topic, providing richer contextual information than machine learning. FMNs build cognitive representations of stances through psycholinguistic tools like conceptual associations from semantic memory (free associations, i.e., one concept eliciting another) and affect norms (valence, i.e., how attractive a concept is). We test FMNs by investigating how Norwegian nursing and engineering students perceived innovation and health before and after a 2-month research project in e-health. We built and analysed FMNs by six individuals, based on 75 cues about innovation and health, and leading to 1,000 associations between 730 concepts. We repeated this procedure before and after the project. When investigating changes over time, individual FMNs highlighted drastic improvements in all students' stances towards "teamwork", "collaboration", "engineering" and "future", indicating the acquisition and strengthening of a positive belief about innovation. Nursing students improved their perception of 'robots" and "technology" and related them to the future of nursing. A group-level analysis related these changes to the emergence, during the project, of conceptual associations about openness towards multidisciplinary collaboration, and a positive, leadership-oriented group dynamics. The whole group identified "mathematics" and "coding" as highly relevant concepts after the project. When investigating persistent associations, characterising the core of students' mindsets, network distance entropy and closeness identified as pivotal in the students' mindsets concepts related to "personal well-being", "professional growth" and "teamwork". This result aligns with and extends previous studies reporting the relevance of teamwork and personal well-being for Norwegian healthcare professionals, also within the novel e-health sector. Our analysis indicates that forma mentis networks are powerful proxies for detecting individual- and group-level mindset changes due to professional growth. FMNs open new scenarios for data-informed, multidisciplinary interventions aimed at professional training in innovation.