Forum Comments Research Articles

Text classification and topic mining of intelligent forum comments in university MOOC based on CNN networks

Text classification and topic mining of intelligent forum comments in university MOOC based on CNN networks

Federated Meta Embedding Concept Stock Recommendation

Mining relevant stocks given a trending topic/concept in capital markets is an application with significant economic and societal impacts. Previous concept stock recommendation system mines concept stocks only from public social media like financial news. On stock forums, investors discuss emerging concepts and stocks by using forum comments, which are unneglectable resources to capture trending concept stocks accurately and timely. However, the comment data from a single forum is insufficient to build a high-quality recommendation system. The forums are data silos protected by privacy regulations, and their comments are still underutilized. In this paper, we propose a federated concept stock recommendation baseline and an optimized method that both leverage the private forum comments and public social media without compromising privacy regulations. Our baseline, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , Federated Meta Embedding (FedME), is built upon the federated learning framework and learns a concept-stock embedding jointly from private and public data. Our optimized method, Federated Graph Meta Embedding (FedGME), improves FedME by using a graph to combine two sources of embeddings and additional human experts' concept-stock knowledge. Empirically, the experiments on two concept stock datasets show that FedME and FedGME substantially improve the performance of recommendation. Our methods provide practical guidance on privacy-preserving FinTech applications.

A Highly Accurate Adverse Drug Reactions (ADR) Detection from Medical Forum Comments Using Long Short-Term Memory Networks

Purpose: Adverse Drug Reactions (ADR) classification is useful in modern medical diagnostics and related applications. ADR is an example of how medical information is frequently accessible on social media platforms for healthcare, where people can share their experiences with treatments on desktop computers and mobile devices. Many researchers are interested in gathering valuable medical data from social media for the ADR system training and classification process. Materials and Methods: This research explores the effects of three aspects on recognizing ADR mentions in social media for the medical field and proposes a deep neural network of Long Short-Term Memory (LSTM) neural networks to do so. The comments are collected from various social media platforms to implement the ADR system with proper training and testing processes. The texts from the dataset are initially preprocessed by using a data filtering and clustering process to remove the input data's redundant information to increase the training process's quality. Characteristic features, such as semantic features and text statistics, are extracted from the input text using the American Standard Code for Information Interchange (ASCII) array. Further, the features are converted and fed to LSTM networks for training and validation. Results and Conclusion: This work is evaluated using two datasets, CODEC, and ADR Corpus datasets are used to evaluate the performance of the proposed ADR technique via multiple angles. Via extensive experiments, this work achieved 99.79 accuracy, 98.37 sensitivity, 97.63 specificity, 99.72 precision, 98.39 recall, 97.62 F1-score for the CODEC dataset, 98.16 for accuracy, 99.19 for sensitivity, 98.49 for specificity, 99.49 for precision, 96.72 for recall, and 93.16 for F1-score for ADR corpus, respectively.

From Conflict to Peace: Fuqaha Political Exegesis on Israel's War Crimes in Palestine

The aims of this research are to analyse the contributions of Indonesian netizens' comments regarding Israel's war crimes in Palestine and to examine the contributions of fuqaha's political exegesis towards peace politics in addressing these crimes. The study provides insights into Indonesian netizens' views and the role of fuqaha in promoting sustainable peace. Using an in-depth qualitative method and textual analysis, the research focuses on fuqaha interpretations of Israel's war crimes, the two-state solution, Palestinian independence, and the role of Islamic law in peace and reconciliation. Participants include fuqaha experts, Islamic studies and international law academics, and Indonesian netizens selected for their expertise. Data collection involves analyzing war crimes documents, two-state resolution texts, and public forum comments. Thematic coding identifies patterns in the data. Findings show Indonesian netizens' strong opinions on social media, calling for accountability for figures like Netanyahu and urging ICC action. Fuqaha, like Al-Shafi'i, Al-Shaukani, and Al-Zuhaily, emphasise peace and justice, condemning Israel's violations of international law. They advocate for integrating Islamic principles into political and social approaches to achieve lasting peace.

Symbolic boundaries and the negative classifications of the ‘Successful Other’ at the intersection of ethnicity, religion, gender, and refugeeness

ABSTRACT Our paper focuses on how members of the national ingroup respond when confronted with ‘successful others’ who challenge prevailing perceptions of immigrants and national belonging. We empirically investigate this through a case study on Alma Zadić, Austria’s Minister of Justice since January 2020. Her non-Austrian origin initiated a heated public debate on the legitimacy of an immigrant or refugee holding one of the country’s most important political positions. We analyse public discourse using interpretive analysis of online newspaper forum comments, employing the symbolic boundaries approach combined with the concept of negative classifications. The results reveal the cultural repertoires that members of the arrival society make use of when they negatively categorise an alleged newcomer of taking up a position that is considered exclusively reserved for the established. We contribute to the cultural sociology of social inequality and advocate for greater consideration of the intersectional entanglement of symbolic boundaries.

The presence of source viewership in fansub paratexts

Abstract Online viewers’ paratextual activities when watching video content include rating, commenting and reviewing. These have received greater attention in audiovisual translation studies, since viewers’ paratexts can provide clues as to how the subtitled product is understood and what elements capture viewers’ attention. This paper examines the English fansubs of a Chinese period drama, to explore Chinese viewers’ motives for watching the show with English subtitles. Special attention is paid to the roles they play in the meaning-making processes of the online community. The paratextual elements include selected timed comments (comments synchronised to the video timeline), forum comments and reviews. The results show that although Chinese viewers are critical of the subtitles, they are highly aware of the fansubbers’ efforts and are enthusiastic about promoting Chinese culture to the wider world. Further, Chinese viewers participate in conversations with non-Chinese online viewers, bringing novel interpretations and circulating creative pleasure.

Bloom-epistemic and sentiment analysis hierarchical classification in course discussion forums

&lt;span lang="EN-US"&gt;Online discussion forums are widely used for active textual interaction between lecturers and students, and to see how the students have progressed in a learning process. The objective of this study is to compare appropriate machine-learning models to assess sentiments and Bloom’s epistemic taxonomy based on textual comments in educational discussion forums. The proposed method is called the hierarchical approach of Bloom-Epistemic and Sentiment Analysis (BE-Sent). The research methodology consists of three main steps. The first step is the data collection from the internal discussion forum and YouTube comments of a Web Programming channel. The next step is text preprocessing to annotate the text and clear unimportant words. Furthermore, with the text dataset that has been successfully cleaned, sentiment analysis and epistemic categorization will be done in each sentence of the text. Sentiment analysis is divided into three categories: positive, negative, and neutral. Bloom’s epistemic is divided into six categories: remembering, understanding, applying, analyzing, evaluating, and creating. This research has succeeded in producing a course learning subsystem that assesses opinions based on text reviews of discussion forums according to the category of sentiment and epistemic analysis.&lt;/span&gt;

Modeling crude oil volatility using economic sentiment analysis and opinion mining of investors via deep learning and machine learning models

Investor emotions can significantly impact the perceived value of financial assets, and sentiment analysis technology can accurately model daily responses and views of investors. In this study, we analyze investor remarks from one of the most popular web economic areas to investigate the effects of investor emotions on the price of crude oil in China. We construct investor emotions indices for China's crude oil futures based on specific economic factors and employ five machine learning technologies, including the AR approach, SVR approach, MLP method, neural networks with a recurrent units model, a recurrent unit with a gate approach, and LSTM model, to predict one-day-ahead prices. Our research offers insights into the effectiveness of different machine learning technologies and how specific economic factors contribute to sentiment analysis accuracy. We discuss the natural language processing techniques and tools used to extract sentiment from financial forum comments, along with the measures taken to ensure the accuracy and reliability of sentiment scoring. Our study highlights the advantages of the LSTM model, paired with the compound sentiment index, in the context of crude oil price forecasting and specifies the accuracy levels achieved by this model in comparison to others. We also discuss potential limitations and challenges in the long-term application of sentiment analysis for price forecasting and propose possible solutions. Our findings have practical implications for investors, financial analysts, and policymakers involved in China's crude oil market. We explore patterns or trends in investor emotions in our analysis and discuss their correlation with major economic or geopolitical events in China and globally. Our study contributes to a deeper understanding of investor emotions' impact on crude oil prices in China and provides valuable insights for various stakeholders in the dynamic and speculative energy oil industry.

“Gua means scrape”: a conversation analysis of identity construction and negotiation in polylogal Wikipedia paratext

ABSTRACT This article contributes to multidisciplinary research exploring the role played by paratext in the construction of identities. First, it develops a theoretical and methodological framework inspired by recent scholarship on conducting conversation analysis in digital contexts and on the ways in which speakers claim positions of expertise through their turns-at-talk. Next, it presents a case study examining translation-focused discussion forum comments posted within the so-called “Talk” pages that surround the mainspace content of the online user-generated encyclopedia Wikipedia. The analysis highlights the often intensely interactive nature of digital paratext and probes the implications for participants’ identity work: I argue that the polylogal nature of this paratextual space requires the constant negotiation of identities as Wikipedia contributors relentlessly jostle with one another for recognition of their epistemic authority. In the final section, I discuss the implications of this Wikipedia case for broader areas of scholarship on translation and digital paratext.

A Dataset of Coordinated Cryptocurrency-Related Social Media Campaigns

The rise in adoption of cryptoassets has brought many new and inexperienced investors in the cryptocurrency space. These investors can be disproportionally influenced by information they receive online, and particularly from social media. This paper presents a dataset of crypto-related bounty events and the users that participate in them. These events coordinate social media campaigns to create artificial "hype" around a crypto project in order to influence the price of its token. The dataset consists of information about 15.8K cross-media bounty events, 185K participants, 10M forum comments and 82M social media URLs collected from the Bounties(Altcoins) subforum of the BitcoinTalk online forum from May 2014 to December 2022. We describe the data collection and the data processing methods employed and we present a basic characterization of the dataset. Furthermore, we discuss potential research opportunities afforded by the dataset across many disciplines and we highlight potential novel insights into how the cryptocurrency industry operates and how it interacts with its audience.

Judging Stock Trends According to the Sentiments of Stock Comments in Expert Forums

Machine learning has been proven to be very effective and it can help to boost the performance of stock price predictions. However, most researchers mainly focus on the historical data of stocks and predict the future trends of stock prices by designing prediction models. They believe that past data must hide useful information in the future. Due to the lack of human participation, the result of this practice must be accidental. To solve this problem, we propose a novel model called Convolutional Neural Network with Sentiment Check (CNN-SC) in this paper. The model recommended by the authors refers to and expands upon the ideas of experts, and then takes the sentiment value in expert comments as the basis for stock price prediction. This model reflects the humanization of stock price prediction and eliminates the problem of a lack of supervision in machine learning. To demonstrate the effectiveness of our novel method, we compare it with five other popular and excellent methods. Although the C-E-SVR&amp;RF and GC-CNN models are also quite effective, our results indicate the superiority of CNN-SC and it is accurately used to calculate the short-term (seven days later) stock price fluctuation of a single stock.

Enhanced Esports: Community Perspectives on Performance Enhancers in Competitive Gaming

This work explores perceptions of performance enhancer usage in esports. Specifically, we explored the perception of: food and food supplements; non-medical use of prescription drugs; drugs with some social acceptance (e.g. alcohol, nicotine, cannabis); drugs with lower social acceptance (e.g., psychedelics, opioids); and non-invasive brain stimulation (e.g. transcranial direct current stimulation). A mixed-methods approach was used to triangulate findings around three data sets, including both prompted and unprompted online forum comments, as well as survey data. The studies evidence that players are willing to use or are already using enhancers to increase their in-game performance, and that players are generally concerned about the use of enhancers in professional esports contexts. Furthermore, the community perceives that a substantial number of e-athletes use enhancers. The core contribution of this work is a comprehensive investigation into perspectives of esports performance enhancement, which highlights the urgent need for further research, as well as regulation by esports leagues.

Exploring Impacts of a Nutrition-Focused Massive Open Online Course.

The nutrition education landscape is changing due to advances in technology. Massive Open Online Courses (MOOCs) are an example of new education opportunities, made possible by advances in online learning environments. This research aimed to evaluate a nutrition-focused MOOC, applying Kirkpatrick’s theoretical model of learning, to comprehensively describe learners’ reactions, knowledge and behaviours. A mixed-methods approach explored learners’ experiences of participating in a global nutrition MOOC. Quantitative survey data, using descriptive statistics, measured pre-course nutrition knowledge, post-course satisfaction and learning, and changes to fruit/vegetable intake. Qualitative data from discussion forums and semi-structured interviews were thematically analysed and explored participant experiences and perceived impacts on dietary behaviours and nutrition knowledge. All results were mapped to Kirkpatrick’s model. Surveys measuring perceived knowledge, satisfaction, and fruit and vegetable intake were completed by 4941, 1003 and 1090 participants, respectively. Overall reactions to the course were positive. Perceived vegetable intake increased for 62% of participants. Twelve interviews and six hundred and forty-seven forum comments were analysed, identifying eight themes highlighting changes to knowledge and dietary behaviours, and the importance of peer-to-peer learning. All results mapped to Levels 1–3 of Kirkpatrick’s model. MOOCs can be an effective platform to communicate evidence-based nutrition information to large, global audiences.

Mining opinions on LMOOCs: Sentiment and content analyses of Chinese students’ comments in discussion forums

The rapid rise of Massive Open Online Courses (MOOCs) has attracted intense discussion on their benefits to education transformation and related challenges. Language MOOCs (LMOOCs) as an emerging field requires more empirical attention, particularly in terms of learners' perceptions and needs regarding LMOOC learning. This study aimed to explore LMOOC students' sentiments and opinions through sentiment analysis and content analysis of their comments posted in the discussion forums of 60 LMOOCs and examine the correlation between student sentiment and course rating. The results indicated that the majority of the comments were positive, and student sentiment was positively correlated with course rating. Five major themes that encapsulated students' opinions about LMOOCs and their experiences in LMOOC learning were identified: attitudes towards the LMOOCs, comments on the LMOOCs, evaluations of LMOOC instruction and instructors, learning outcomes, and suggestions. To shed light on LMOOC development, the LMOOC learners’ concerns and suggestions were particularly examined, and tactics possibly used by learners to express euphemistically or circumvent negative feedback were speculated. Implications for LMOOC instructors and developers and future research are finally presented.

Homophobia in Brazilian football: A critical discourse analysis of fans’ comments in online football forums

Football is one of the many fields where the display of orthodox masculinity, often related to sexism and homophobia, reaches its peak. Studies indicate that such behaviors negatively affect athletes who do not fit heteronormative standards and closeted gay players, who fear coming out due to an intimidating and toxic environment. Fans have been central to some empirical studies investigating homophobia in football; however, most research to date focuses on western countries. Through the lens of Critical Discourse Analysis, this paper examines how Brazilian football fans see the presence of openly gay players on a men's team. One hundred and fifty comments were collected from thirteen online football forums. The results indicate similarities in the responses provided by Brazilian fans compared to western studies. Although fans’ positive views towards the presence of openly gay players were expressed, homophobic opinions were also identified in this environment. Several fans also demonstrated no objection to the use of homophobic language. The present study is relevant to the existing literature as it aids in clarifying previous notions of homophobia in Brazilian football. At the same time, it problematizes and officially places Brazil on the map of studies on the topic.

Another victory of retail investors: Social media's monitoring role on firms' earnings management

Stock forums are expected to complement individual investors' sophistication via collective wisdom, thus triggering extreme market reactions through uniform scattered opinions. This paper investigates whether stock forums improve corporate governance and decrease earnings management among firms. We found that increased posts, views and comments in a stock forum raise firms' potential costs for obfuscating corporate performance and constrain their upward earnings management. Stock forum discussion tends to increase the likelihood of and the penalties for earnings management detection. By decomposing the cost of earnings management into tangible and regulatory costs, we find that the stock forum governance effect is more pronounced in firms with higher managerial ownership and higher regulatory risk. Social media is a complement and intensifier of traditional mass media. Firms with broader media coverage are more sensitive to the stock forum governance effect. Our results are robust after considering the likelihood that executives will account for stock forum discussion in their reporting decisions.

Alternative Opinions about Vaccination in Religious Discourse

This article examines speech strategies and tactics in religious communication in the context of the coronavirus pandemic. The purpose of the work is to consider the features of the speech behaviour of communicants during the pandemic crisis. The research methodology is based on the theory of speech communication to some extent present in modern works on the study of discourses of various types, the study of the peculiarities of religious communication, communicative/speech strategies, and tactics applied to different speech material. The researcher refers to texts published in Orthodox media and on Orthodox websites: sermons, comments in Orthodox forums, and posts in social networks. As a result, the author reveals the polemical nature of this segment of religious discourse. For instance, it contains neologisms with evaluative connotations, such as цифровой концлагерь (digital concentration camp), псевдопандемия (pseudo-pandemic), коронобесие (corona-frenzy), коронокризис (corona-crisis), ковид-диссиденты (covid-dissidents), антипрививочники (anti-vaccinators), антиваксеры (antivaxers), ковидиоты (covidiots), антивакс-сектанты (anti-vax sectarians), ковидофобы (covidophobes), вакциноборцы (vaccine fighters), вакцина каннибала (cannibal vaccine), вакциноскептики (vaccine sceptics), etc. The author characterises two types of addressees: 1) those denying the existence of the pandemic and/or the need for preventive measures and/or vaccination; 2) those recognising the fact of the pandemic and/or the need for preventive measures and/or vaccination. The former are characterised by a confrontational type of communication, as they choose strategies of discrediting and manipulation by means of accusations, insults, slander, sarcasm, appeals to “authority”, etc. They violate the principles of cooperative communication and show verbal aggression. The various phobias that the communicants of this group demonstrate are due to a misunderstanding of the soteriological and eschatological teachings of the Church. The addressees of the second group use a speech strategy of persuasion through interpretation, explanation, references to their own experience, and the authority of the Church. The communicants of this group obviously aim for conflict-free communication.

Spread of gambling abstinence through peers and comments in online self-help chat forums to quit gambling

Habit formation occurs in relation to peer habits and comments. This general principle was applied to gambling abstinence in the context of online self-help forums to quit gambling. Participants in this study, conducted between September 2008 and March 2020, were 161 abstinent and 928 non-abstinent gamblers who participated in online self-help chat forums to quit gambling. They received 269,317 comments during their first 3 years of forum participation. Gamblers had an increased likelihood of 3-year continuous gambling abstinence if they had many peers in the forums. However, they had a decreased likelihood of gambling abstinence if they received rejective comments from the forums. Based on these results, online social network-based interventions may be a new treatment option for gamblers.

Humorous effect when linguocultural phenomena are included in forum comments to media texts

The paper is devoted to the possibilities of creating a comic effect based on linguocultural signs and concepts in forum comments to media texts. The problem of the comic is studied in the context of the study of value essences and national-cultural meanings, which form the basis of the worldview of the national-cultural community. The material of the study is forum comments on Bulgarian online media publications or reposts of media articles in social networks. Due to the fact that the specificity of modern online communication is the relativity of the disposition of the author-addressee, the addressee is involved in the discussion of the topic, since both media texts and comments on them are usually read. Forum comments as posts written by the addressee under media texts, create a kind of network communication that allows its participants to evaluate the news or express an opinion about the content of the media text and, in some cases, to simply argue with or without reason. The formation of a comic effect on the basis of linguistic and cultural phenomena in forum communication is an important linguo-mental characteristic of the perception of the world by a national-cultural community, as well as a tool for assessing the preservation of cultural tradition. The analysis of linguocultural material involved in creating a comic effect in comments to media texts demonstrates the presence of several types of precedent, the use of phraseological units (paremias, citations, aphorisms), metaphors and concepts. The “transposition” of linguistic and cultural phenomena is based on the discrepancy between the internal and external image, on the violation of the usual syntagmatic relations of linguistic units, the actualization and modification of frame-slot connections, and the ambivalence of semantic realization. The article is based on the report that is going to be presented at the XVIII International Congress of Slavists in 2025.

Judging Stock Trends According to the Sentiments of Stock Comments in Expert Forums

Judging Stock Trends According to the Sentiments of Stock Comments in Expert Forums