Opinion Orientation Research Articles

POLOR: Leveraging Contrastive Learning to Detect Political Orientation of Opinion in News Media

News articles are naturally influenced by the values, beliefs, and biases of the reporters preparing the stories and the policies of the publishing outlets. Numerous studies and datasets have been proposed to detect the political orientation of news articles. However, most of these studies ignore real textual clues and learn the textual signature of the source (commonly the publisher and rarely the writer) of the article instead. Moreover, a good volume of opinion pieces published by major news outlets do not reflect the political orientation of the publisher but rather reflect the political orientation of a non-professional writer. Existing methods are not built to correct this difference in the training data and, hence, perform poorly on human-annotated data. We propose, POLOR, a fine-tuned BERT model that employs contrastive learning to detect the political orientation of news articles even when the training data is labeled by the source (i.e. the publisher of the news article). Unlike previous work in the literature, the model learns features by employing different contrastive learning objectives where each sentence is contrasted with sentences from various sources simultaneously. POLOR achieves a 15\% increase on our dataset compared to previously proposed baselines. Finally, we release two datasets of opinion news: source-annotated and human-annotated datasets. The full paper including supplementary materials, code, and datasets can be found at https://www.cs.unm.edu/~ajararweh/.

Consumers’ Opinion Orientations and Their Credit Risk: An Econometric Analysis Enhanced by Multimodal Analytics

Consumers’ Opinion Orientations and Their Credit Risk: An Econometric Analysis Enhanced by Multimodal Analytics

Research on the Orientation of Public Opinion by New Media: Taking the Jiang Ge Case as an Example

New media plays an increasingly important role in peoples lives because of its various forms, wide coverage, and strong interaction. While giving birth to a new type of media platform, it also makes the public a disseminator of public opinion. The role of new media should not be underestimated. This paper takes the Jiangge Case as an example and combines the existing literature and relevant data to study the new medias opinion orientation, and further analyzes the strategy of the new medias public opinion guidance. Through exploring the influence of new medias public opinion guidance on the Jiang Ge case, it is concluded that the publics emotion will make the publics opinion out of control, and the new media platform will play an indispensable role in public opinion guidance. New media should deepen their professionalism, correctly guide opinion leaders in the dissemination of speech, and at the same time establish a good mechanism for the dissemination of public opinion to prevent the generation of inappropriate speech from the root, thus avoiding the uncontrollable development of information dissemination.

Why China Cares about Canada’s Indigenous Residential Schools: from Whataboutism to Internal Denial

This article examines how the Chinese government and its propaganda departments use genocide-related discourses to fulfil different political purposes at home and abroad. By criticizing Western colonialist regimes’ assimilation policies, especially Canada’s Indigenous residential schools, the Chinese diplomats apply the rhetoric of whataboutism to dodge the international community’s questions about China’s systematic persecution of Uyghur Muslims. Domestically, China’s state media intensively cover Canada’s residential school system and the colonial genocide against Indigenous people, trying to distract the audience from the state atrocities in Xinjiang and mislead the public to distrust Canada and other countries’ motives for accusing China of committing genocide. This media campaign is an example of the Chinese government’s “ yu lun dao xiang” (public opinion orientation) propaganda. It deploys a mirroring strategy that makes use of the agendas related to the genocides in other countries to cover up a genocide that is happening at home; it turns the residential school survivors’ trauma into a tool to defend a genocidal system that likewise takes the form of education. This strategy poses a new challenge to the diplomatic and academic work that aims to prevent genocide.

Implementation Opinion Mining For Extraction Of Opinion Learning In University

Opinion mining is a field of Natural Language Processing (NLP) that is used to carry out the process of extracting and processing textual data which functions to obtain information through sentiment analysis from a document in the form of text, among others, to detect attitudes towards objects or people. Sub-processes in opinion mining can use documents of subjectivity, opinion orientation, and detection targets to find out the data used as sentiment analysis, sentiment analysis aims to assess emotions, attitudes, opinions, and evaluations conveyed by a speaker or writer towards a product or towards a public figure. In this study, an opinion mining system was developed to analyze learning in college. The methodology used is quantitative descriptive, while the processing of sentiment analysis data uses Azure machine learning. Sentiment analysis results are very good at assessing opinions or opinions and emotions, and attitudes conveyed by someone. The learning process is the main element that must run well so that competency and achievement in learning can be maximally conveyed to students. Documents that identified opinions were then classified into negative, neutral, and positive opinions based on the results. In general, it can be concluded that the value obtained by sentiment analysis using Azure Machine Learning tools is quite good, judging from the results of a positive class of 0.79 and a neutral class of 0.53. The use of cleaning and labeling techniques and other classifications is still very possible to use. To get a better accuracy value.

Politics, publicity and social movements in the works of J. Habermas and P. Bourdieu

Modern social movements should be considered in the context of the existing public sphere, the main theorists of which are J. Habermas and P. Bourdieu. The article analyzes various aspects of the functioning of the public sphere, examines the role of mass media in the formation of the public sphere and the orientation of public opinion. Using the example of social movements in France, the author examines how the public sphere affects the functioning of social movements. Nowadays, the media largely determines the agenda and controls public debate. Besides, the experience of each person is completely mediated by the communication means. At the same time the media is largerly influenced by politics and, as a result, the public sphere loses its functions. Nevertheless, the public of organized individuals could become an effective participant in public communication, through which it is possible to legitimize the achievement of political compromises. The author concludes that social movement's analysis should be focused not only on the political, economic and social causes of demonstrations, but also on the public sphere in which they exist. In addition, the discourse constructed by the mass media is not only imposed on society, but belongs to it. Each person contributes to its creation, and social media platforms play an important role in this. If public discourse is the result of the activity of many actors, then it is necessary to consider the possibility of interference in its construction.

Hot-topics cross-propagation and opinion-transfer dynamics in the Chinese Sina-microblog social media: A modeling study

On social media platforms, hot topics often contain several pieces of related information that can influence internet users, generating either positive or negative opinion orientation. Some of them will choose to retain or change their original opinions after exposure to multiple related messages. To describe the opinion-transfer transient and collective behaviors in this scenario, this paper proposes an opinion-transfer susceptible-forwarding-immunized (OT-SFI) information cross-propagation model. Real multiple information in messages with opinions obtained from the Chinese Sina microblog is used for data fitting to illustrate how model parameters can be estimated and used to predict the accumulative numbers of users with a particular view. The study attempts to relate changes in group views in the network to initial opinion distribution and individuals' opinion choices at the macro level. Furthermore, the model parameters at the micro level are used to measure the probability of “retention” and “reversal” of views in events, as well as the extent to which the masses are influenced by new information views. The result illustrates that the viewpoint distribution of the initial message and the opinion selection of the new message opinion leaders play crucial roles in promoting attention to the topic and driving for a desired collective opinion.

Polluting the Political Discourse

Abstract Modern deliberative democracies are built to ensure pluralism in the political discourse so that citizens are able to express informed preferences when taking part in public policymaking by electing their representatives. The fairness of political processes is threatened in the digital environment by the resort to micro-targeting techniques and the spread of disinformation campaigns, which may poison the authenticity of public debate and influence the orientation of public opinion on policy issues. Assessing the role of digital intermediaries in these dynamics is therefore critical for European countries and EU institutions in order to ensure the survival of the paradigm of pluralistic democracy. The trade-offs between State regulation of freedom of expression online, platforms’ accountability and self-regulation make it compelling to analyse whether EU policy choices are adequate to confront phenomena undermining democratic processes.

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

SummaryCross‐domain sentiment classification is a significant task of sentiment analysis that objectives to predict the opinion orientation of text documents in the target domain by using the source domain's learned classifier. Most of the existing approaches of domain‐adaptation in sentiment classification focus on sharing low‐dimensional features across the domain using domain independent and specific features to mitigate the gap between domains. Earlier cross‐domain sentiment classification approaches mainly focused on document level and sentence level, they cannot consider the full impact of aspect words, position of the words, and long‐term dependencies. To address this concern, we propose a model for cross‐domain sentiment classification, which is based on decoding‐enhanced BERT with disentangled attention (DeBERTa). DeBERTa is a pretrained language model based on transformer architecture. In this article, we perform sentence and aspect embedding to mine wordpiece information from text document. DeBERTa language‐model utilize disentangled attention mechanism and an enhanced mask decoder to understand the expression features. Disentangled attention mechanism is used to encode each word into two vectors (i.e., content and position vector). In order to predict the masked tokens during model pretraining, an enhanced mask decoder is employed, which incorporates absolute positions in the decoding layer. Finally, experiments are conducted on the benchmark dataset that demonstrates the superiority of fine‐tuned DeBERTa model for cross‐domain sentiment classification tasks.

Evaluating Online Products Using Text Mining: A Reliable Evidence-Based Approach

To address the uncertainty about the quality of online merchandise, e-commerce sites often provide product review ranking services to help customers make purchasing decisions. Such services can be very useful, but they are not necessarily reliable when the ranking results are based on ratings without considering their reliability. In this paper, we propose a reliable evidence-based approach to online product evaluation by using text mining to analyze product reviews while taking into account the reliability of each review. We parse the product reviews and classify the opinion orientations for each recognized product feature as positive or negative. Then, we weigh the classified opinion orientations by their reliability and use them as independent evidence to calculate the belief values of the product using Dempster–Shafer (DS) theory. Based on the belief values of a list of similar products, we can calculate their product effectiveness and cost-effectiveness values for product ranking. The case studies show that our approach can greatly help customers make better decisions when choosing the right online products.

DRMM: A novel data mining-based emotion transfer detecting method for emotion prediction of social media

With the progress of the Internet and information technology, emotion analysis has been applied to analyse the emotional orientation and evolution trend of online public opinion of online tweets. At present, most of the existing methods use econometric model and machine learning algorithm to predict the trend of online public opinion. Although these methods have achieved good prediction results, they do not take into account the influence of internal factors on network public opinion prediction, such as mutual migration among emotion classes. The emotion may change dynamically because different events trigger it in the evolution process. In this view, this article proposes a novel method, called Deviation Rule Markov Model (DRMM), to predict the emotional change trend of Internet users in online public opinion by analysing the correlation between Internet users’ emotional categories. Structurally, the proposed DRMM involves various processes such as pre-processing, emotion classification, data mining and transfer prediction. For the processing of network comment data, the proposed model initially undergoes pre-processing to delete unnecessary data. Then, the extended fuzzy emotion ontology is used to annotate the emotion class of the comment data. Besides, an extended association rule mining algorithm is used in the emotion association analysis process to obtain the transfer probability between emotion classes. Moreover, Markov chain is used to construct an emotional state transition matrix to predict the transition probability of positive or negative emotions. According to the predicted single emotion transfer probability results, the analytic hierarchy process is used to assign values to different emotion classes, and finally, the transfer probability of the overall emotion in a certain period is obtained. Compared with the actual case, the mean absolute error (MAE) and root mean square error (RMSE) of the proposed model are 2.7119 and 3.7254, respectively, which has good prediction performance.

The regulatory power of social expectations: developing a measurement scale

PurposeThis research is devoted to the study of social expectations, and the purpose of this paper is to elaborate the instrument of their measurement based on the main spheres of influence.Design/methodology/approachThe research was conducted using a mixed methodology: a series of semistructured interviews and a survey. In the first stage, 15 interviews, and in the second phase, both online and offline surveys (306 participants) were conducted to test the social expectations scale designed. The qualitative phase highlighted the most important areas of social expectations, identifying how they can influence individual behavior. Afterward, the scale of social expectations was developed, and its prognostic function was confirmed.FindingsIt was revealed that expectations influence goal setting, motivation, public opinion orientation, emotional experience and decision-making.Originality/valueThis study is an attempt to construct a measurement tool for social expectations and close the gap for many studies that used the concept without operationalization.

Mixed frequency composite indicators for measuring public sentiment in the EU

Monitoring the state of the economy in a short time is a crucial aspect for designing appropriate and timely policy responses in the presence of shocks and crises. Short-term confidence indicators can help policymakers in evaluating both the effect of policies and the economic activity condition. The indicator commonly used in the EU to evaluate the public opinion orientation is the Economic Sentiment Indicator (ESI). Nevertheless, the ESI shows some drawbacks, particularly in the adopted weighting scheme that is static and not country-specific. This paper proposes an approach to construct novel composite confidence indicators, focusing on both the weights and the information set to use. We evaluate these indicators by studying their response to the policies introduced to contain the COVID-19 pandemic in some selected EU countries. Furthermore, we carry out an experimental study where the proposed indicators are used to forecast economic activity.

Psychiatry on Twitter: Content Analysis of the Use of Psychiatric Terms in French.

BackgroundWith the advent of digital technology and specifically user-generated contents in social media, new ways emerged for studying possible stigma of people in relation with mental health. Several pieces of work studied the discourse conveyed about psychiatric pathologies on Twitter considering mostly tweets in English and a limited number of psychiatric disorders terms. This paper proposes the first study to analyze the use of a wide range of psychiatric terms in tweets in French.ObjectiveOur aim is to study how generic, nosographic, and therapeutic psychiatric terms are used on Twitter in French. More specifically, our study has 3 complementary goals: (1) to analyze the types of psychiatric word use (medical, misuse, or irrelevant), (2) to analyze the polarity conveyed in the tweets that use these terms (positive, negative, or neural), and (3) to compare the frequency of these terms to those observed in related work (mainly in English).MethodsOur study was conducted on a corpus of tweets in French posted from January 1, 2016, to December 31, 2018, and collected using dedicated keywords. The corpus was manually annotated by clinical psychiatrists following a multilayer annotation scheme that includes the type of word use and the opinion orientation of the tweet. A qualitative analysis was performed to measure the reliability of the produced manual annotation, and then a quantitative analysis was performed considering mainly term frequency in each layer and exploring the interactions between them.ResultsOne of the first results is a resource as an annotated dataset. The initial dataset is composed of 22,579 tweets in French containing at least one of the selected psychiatric terms. From this set, experts in psychiatry randomly annotated 3040 tweets that corresponded to the resource resulting from our work. The second result is the analysis of the annotations showing that terms are misused in 45.33% (1378/3040) of the tweets and that their associated polarity is negative in 86.21% (1188/1378) of the cases. When considering the 3 types of term use, 52.14% (1585/3040) of the tweets are associated with a negative polarity. Misused terms related to psychotic disorders (721/1300, 55.46%) were more frequent to those related to depression (15/280, 5.4%).ConclusionsSome psychiatric terms are misused in the corpora we studied, which is consistent with the results reported in related work in other languages. Thanks to the great diversity of studied terms, this work highlighted a disparity in the representations and ways of using psychiatric terms. Moreover, our study is important to help psychiatrists to be aware of the term use in new communication media such as social networks that are widely used. This study has the huge advantage to be reproducible thanks to the framework and guidelines we produced so that the study could be renewed in order to analyze the evolution of term usage. While the newly build dataset is a valuable resource for other analytical studies, it could also serve to train machine learning algorithms to automatically identify stigma in social media.

Quantitative Evaluation and Optimal Decision-Making Study on the Cultivation under the Environment of Humanistic Literacy Based on Advanced Data Analysis Technologies.

As an important part of China's higher education, higher vocational education occupies an extraordinarily important position in China's higher education system. In order to clarify the literacy of higher vocational students, carry forward their excellent quality, and correct their bad habits, a quantitative evaluation and optimization decision-making research on the cultivation of humanistic literacy of higher vocational students based on data mining technology is proposed. By using data mining technology, the important data of students can be obtained quickly; by using advantages of information integration, the cultivation of humanistic quality of higher vocational students is evaluated from a correct perspective; by strengthening the education of campus culture construction, students' moral quality can be enhanced, students' enthusiasm for learning can be mobilized, the advantages of new media's public opinion orientation can be given full play, and a humanistic and healthy growth environment for vocational students can be created. Through experiments, it is proved that data mining can better evaluate the humanistic quality of students. During the learning process, the humanistic quality evaluation of female college students reaches 95%, and the overall quality of male and female gradually rises to 80%.

Opinion Mining of Environmental Issues on Instagram

Indonesia has a complicated waste issue and is one of the biggest waste producers worldwide. The development of Instagram as one of the popular social media in conveying public opinion in the form of photos and captions has the potential to produce useful information for solving the waste issue. The study aims to investigate public opinion particularly the Instagram users about the waste crisis. It applied opinion mining in which the data sets were obtained from Instagram posts. The overall number of posts using #sampah until December 20th, 2019 was 109K, and the research sample cover posts from January 1st, 2018 to December 20th, 2019, totaling 54K posts. The developed system consists of three subprocess, namely document subjectivity, opinion orientation and target detection. The data were classified using feature reduction. The result showed that the most popular type of post was about social learning, while the meme posts were relatively low. The conclusion stated that public opinion on the waste issue in the captioned photos can be categorized into five including waste crisis monitoring, social learning, information enlightenment, cultural transformation, and memes. The discussion and solutions elaborated in this study can be utilized by the stakeholders of Penta Helix to realize the solution for waste issues

On the Features of Contemporary Youth’s Values on Life and Educational Countermeasures

With an increasing development and boom of China’s socialist market economy, people’s living standards have been greatly improved and promoted. While conducting economic activities and trading with other countries, we are also actively carrying out cultural exchanges. As a result, people’s traditional ideologies have been impacted to various degrees, and their values have also been changed greatly. This change has a positive and negative impact on the formation of the values of life of young people who are actively thinking and not deeply involved in the world. The life values of contemporary adolescents are becoming more diversified, and there has been a deteriorating trend, especially when it comes to dealing with the issue of “the relationship between dedication and demand”.This paper mainly expounds the characteristics of teenagers’ life values, and puts forward relevant scientific and reasonable educational countermeasures: in view of the current situation of teenagers’ life values, we should carry out value education for contemporary teenagers, strive to establish a correct orientation of public opinion, and hope to have a certain reference and reference for the formation of teenagers’ life values in our country.

A New Neural Networks-Based Integrated Model for Aspect Extraction and Sentiment Classification

The aspect-based sentiment analysis (ABSA) task consists of two closely related subtasks: aspect extraction and sentiment classification. However, the majority of previous studies looked into each task separately, limiting their effectiveness. In contrast, the integration of aspect extraction and sentiment classification into a single model improves results. The main focus in this work is to manage these two tasks into a new collapsed model. The proposed model relies upon the bidirectional long short-term memory (Bi-LSTM) architecture. On the one hand, it combines a multi-channel convolution layer with an optimization method for handling the aspect extraction task. On the other hand, it includes an attention mechanism based on the residual block and aspect position information for predicting the appropriate opinion orientation of an aspect. The experimental results demonstrate that the model achieved the best performance.

Sentiment Weighted Word Embedding for Big Text Data

Sentiment analysis is the practice of eliciting a sentiment orientation of people's opinions (i.e. positive, negative and neutral) toward the specific entity. Word embedding technique like Word2vec is an effective approach to encode text data into real-valued semantic feature vectors. However, it fails to preserve sentiment information that results in performance deterioration for sentiment analysis. Additionally, big sized textual data consisting of large vocabulary and its associated feature vectors demands huge memory and computing power. To overcome these challenges, this research proposed a MapReduce based Sentiment weighted Word2Vec (MSW2V), which learns the sentiment and semantic feature vectors using sentiment dictionary and big textual data in a distributed MapReduce environment, where memory and computing power of multiple computing nodes are integrated to accomplish the huge resource demand. Experimental results demonstrate the outperforming performance of the MSW2V compared to the existing distributed and non-distributed approaches.

Hesitancy toward COVID-19 vaccines among medical students in Southwest China: a cross-sectional study

ABSTRACT Vaccine hesitancy is an obstacle to COVID-19 vaccination. This study aims to inform the vaccine hesitancy and analyze related factors toward COVID-19 vaccination among medical students in China, so as to provide suggestions for increasing vaccines uptake. An online cross-sectional survey was conducted among medical students in a medical university and a health school. A total of 2,196 valid questionnaires were received. 41.2% vaccine hesitancy was reported among medical students in total. Female (OR = 1.336) and individuals with higher education (OR = 1.724) reported slightly higher vaccine hesitancy. Being in low-risk areas and no need to get vaccinated (OR = 2.285), fear of serious consequences of vaccination (OR = 1.929), being in good health and no need to be vaccinated (OR = 1.891), being concerned about short-term side effects (OR = 1.793) and being concerned that the vaccine was ineffective (OR = 1.694) had higher prevalence of vaccine hesitancy. Those who were believing the specialty of study or work environment made vaccination necessary (OR = 0.378), believing they were susceptible to COVID-19 (OR = 0.411) and the vaccine was free (OR = 0.519) were more willing to get vaccinated. Confidence in vaccines and perceptions of benefits and risk are associated with vaccine hesitancy. It is important to strengthen vaccine health literacy education for medical students and enhance vaccine confidence. Countries need to supervise the public opinions in social media, television broadcasting and other media, so as to ensure the correct orientation of public opinion. Open and transparent evidence-based information is also needed which can help improve the vaccination coverage rate of the public.