Social Media Comments Research Articles

Recognizing important factors of influencing trust in O2O models: an example of OpenTable

Online-to-offline/offline-to-online (O2O) business models have attracted lots of enterprisers to enter this market. In such a fast-growing competition, some studies indicated that lack of trust will bring a great damage to O2O business. Related works already confirm trust is the key factor to the success of O2O. Besides, social media has been changing the way providers communicate with consumers. Negative comments in social media will decrease the consumers’ trust to O2O companies and platforms. Available O2O studies are almost always conducted by means of questionnaires or interviews, which cannot provide immediate customer response and require a lot of manpower and time. Since online reviews are the main information source for consumers. Therefore, this study presented a text mining-based scheme which uses text mining technique to find important factors from online electronic word-of-mouth, to replace the traditional questionnaire survey method of collecting data. Two feature selection methods, Support Vector Machines Recursive Feature Elimination and Least Absolute Shrinkage and Selection Operator have employed to select important factors that affect O2O trust. We also evaluate the performance of extracted feature subsets by Support Vector Machines. The findings can be referenced for O2O market enterprises to carefully response customers’ comments to avoid hurting customers’ trust and improve service quality.

The serious side to funny cartoons: understanding public perception of wine through cartoon content analysis

ABSTRACTUnderstanding public perception of a wine festival, organic wine, or the impact of climate change on wine quality can be a complex task. Wine consumers’ opinions, thoughts, feelings and attitudes seem to appear in traditional channels, such as newspapers and magazines, as well as in digital channels, such as blogs, tweets, text messages, social media comments, and consumer ratings. These come in all sorts of formats, but most commonly through text (e.g. posts, tweets) and images (e.g. pictures and videos). Content analysis can be an effective way to understand these widely shared means of expressing sentiment towards a wine and the wine industry. This article examines 300 wine cartoons using a content analysis method that classifies their content into four analytical dimensions: narrative, domestication, binary struggle, and normative transference. This cartoon content analysis reveals details of how different types of wine customers consume and evaluate wine across contexts (e.g. wine in restaurants or at home). This analysis also explores public perception trends regarding wine: social status associations, emotional consumer responses, and consumption-specific concerns. We conclude by discussing future research directions and managerial implications.

Comments Mining With TF-IDF: The Inherent Bias and Its Removal

Text mining have gained great momentum in recent years, with user-generated content becoming widely available. One key use is comment mining, with much attention being given to sentiment analysis and opinion mining. An essential step in the process of comment mining is text pre-processing; a step in which each linguistic term is assigned with a weight that commonly increases with its appearance in the studied text, yet is offset by the frequency of the term in the domain of interest. A common practice is to use the well-known tf-idf formula to compute these weights. This paper reveals the bias introduced by between-participants’ discourse to the study of comments in social media, and proposes an adjustment. We find that content extracted from discourse is often highly correlated, resulting in dependency structures between observations in the study, thus introducing a statistical bias. Ignoring this bias can manifest in a non-robust analysis at best and can lead to an entirely wrong conclusion at worst. We propose an adjustment to tf-idf that accounts for this bias. We illustrate the effects of both the bias and correction with with seven Facebook fan pages data, covering different domains, including news, finance, politics, sport, shopping, and entertainment.

The use of Facebook in the recruitment of foster carers: A dialogic analysis

Social media is becoming increasingly important for communication and community building, yet research on the use of social media by non‐profit organisations is limited and largely restricted to content analysis of social media comments. This article contributes to addressing this research gap, through a survey‐based study of the perspectives of key informants in U.K. Local Authority fostering teams on their use of social media. Specifically, it examines the extent to which the Facebook activity of local authority fostering teams is aligned with the principles of successful social media engagement, as represented by dialogic strategies and outcomes. A questionnaire on the use of Facebook was circulated to all local authority fostering teams in England. Findings suggest that although there is progress, many teams are at an early stage in their social media journey and that there is considerable variation between agencies. The limited evidence of engagement in relation to dialogic principles suggests that there is some adoption of a strategic approach. In particular, of the three dialogic principles associated with successful online engagement, two (updating and community building) were applied by about half of local authority fostering teams and the third (engagement) by just over a quarter.

Online Blaming and Intimate Partner Violence: A Content Analysis of Social Media Comments

Social media has become a ubiquitous form of interacting and sharing information. However, comments on social media sites are often aggressive and contemptuous, especially when topics are controversial or politically charged. For example, discussion of intimate partner violence (IPV) tends to provoke strong reactions from outsiders, who make angry or blaming remarks about those involved. Although IPV is common, it has not been widely discussed in popular media until recent years when high-profile cases of abuse have come to light. In 2016, a celebrity accusation of domestic violence led to thousands of comments on social media, with outsiders weighing in about who was at fault and what should be done. This study involved a content analysis of 400 of these comments, with the intent of better understanding typical types of social media reactions to domestic violence accusations. Key themes included judgment and blame, with around 37% of commenters blaming the supposed victim in this case, while only 9% blamed the alleged perpetrator. The findings show how people comment about domestic violence and illustrate the contentious and often distorted nature of social media interactions. Implications for professionals and researchers are discussed.

Multi-Objective Ant Colony Optimization for Automatic Social Media Comments Summarization

Summarizing social media comments automatically can help users to capture important information without reading the whole comments. On the other hand, automatic text summarization is considered as a Multi-Objective Optimization (MOO) problem for satisfying two conflicting objectives. Retaining the information from the source of text as much as possible and producing the summary length as short as possible. To solve that problem, an undirected graph is created to construct the relation between social media comments. Then, the Multi-Objective Ant Colony Optimization (MOACO) algorithm is applied to generate summaries by selecting concise and important comments from the graph based on the desired summary size. The quality of generated summaries is compared to other text summarization algorithms such as TextRank, LexRank, SumBasic, Latent Semantic Analysis, and KL-Sum. The result showed that MOACO can produce informative and concise summaries which have small cosine distance to the source text and fewer number of words compared to the other algorithms.

From war journalism to peace journalism: Re-inventing peace journalism through audience oppositional reading of terrorism news online

Coverage of terrorism can be categorized into either peace journalism or war journalism. Peace journalism highlights peace initiatives and tones down differences, while war journalism emphasizes differences, often legitimizing violence as the means to peace. This article analyses the war/peace journalism binaries by newspapers in reporting the 2015 terror attack on Kenya’s Garissa University. The article considers an oppositional reading of mainstream media through social media comments. By allowing opposition reading, social media enables the audience to challenge the war journalism that is prominent in mainstream media reporting of terrorism. While war journalism legitimizes violence as the key counter-terrorism strategy, peace journalism seeks to reveal the root causes of conflict. Although the concept of peace journalism is well grounded, its application to oppositional readings of terrorism news remains under-theorized because terrorism is still outside the bounds of what is considered legitimate war. Reporting from a war journalism perspective, the mainstream media focused on coverage that delegitimized terrorism and paid scant attention to alternative understandings of the event. . Contrary, comments on social media discussed terrorism from a peace journalism perspective, focusing on the root cause as well as possible solutions. As part of peace promotion efforts, this study proposes that journalists should be critical of the dominant knowledge structure and report terrorism from a peace journalism perspective. Key words: Al-Shabaab, Peace journalism, terrorism, war, violence, media, news, social media, oppositional reading.

What to Believe? Social Media Commentary and Belief in Misinformation

Americans are increasingly turning to social media for political information. However, given that the average social media user only clicks through on a small fraction of the political content available, the brief article previews that appear in the News Feed likely serve as shortcuts to political information. Yet, in addition to sharing political news, social media also allow users to make their own comments on news posts, comments which may challenge or distort the information contained in the articles. In this paper, we first analyze how social media posts on Twitter and Facebook differ from the actual content of their linked news articles, finding that social media comments regularly misrepresent the facts reported in the news. We then use a survey experiment to test the consequences of these information discrepancies. Specifically, we randomly assign individuals to read a full news article, a news article preview post (as seen on Facebook), or a news article preview with misinformative social commentary attached. We find that individuals in the social commentary conditions are more misinformed about the featured topic, tending to report the factually-incorrect information relayed in the comments rather than the factually-correct information embedded within the article preview.

Customer Reviews: Motivated by Error-Free Service or Successful Recovery?

Across two studies, customer perceptions and outcomes of quality service in different industries were investigated using a qualitative approach that involved computer-based text analysis. In study 1, 60 participants were asked to provide positive and negative experiences in the restaurant and cell phone industries. In study 2, 900 social media comments were collected regarding the cell phone, airline, and restaurant industry. Perceptions of quality service were not necessarily error free. Error free service was, however, expected in some industries. Future research should continue to mine text data to better understand the customer experience.

The national, regional and city divide: Social media analysis of stakeholders views regarding biological controls. The public reaction to the carp control herpes virus in Australia

Recent research and practice in environmental management suggest sentiment analysis of social media communication can be a useful tool in stakeholder analysis of environmental policy. This is certainly the case when it comes to the controversial use of biological controls in dealing with invasive species. Current numerical scored approaches of sentiment may not reveal the reasons for support or opposition to environmental policies in this and other areas. In this study, we examine how the use of more in-depth analysis based on what key stakeholders say about this issue in media at a national, city or regional geographic level. The analysis reveals quite different reasons for support and opposition to the biological control of carp in Australia, and that within each stakeholder group it is possible for individuals to hold conflicting views and attitudes on this issue. We find that there are concerns at the national and city level about the impact of the virus and mistrust of government and science and that the carp species may be viewed as a valuable resource. Those responding to regional media expressed hope that the virus may lead to the elimination of the carp problem, however, they were more interested in the possible impact on the local environment. The multi-scaled social media analysis of stakeholder views about the potential biological control of carp in Australia demonstrated how social media comments can be used to explore the nuanced and multidimensional nature of community attitudes and preferences.

A survey on sentiment classification algorithms, challenges and applications

Abstract Sentiment classification is the process of exploring sentiments, emotions, ideas and thoughts in the sentences which are expressed by the people. Sentiment classification allows us to judge the sentiments and feelings of the peoples by analyzing their reviews, social media comments etc. about all the aspects. Machine learning techniques and Lexicon based techniques are being mostly used in sentiment classification to predict sentiments from customers reviews and comments. Machine learning techniques includes several learning algorithms to judge the sentiments i.e Navie bayes, support vector machines etc whereas Lexicon Based techniques includes SentiWordnet, Wordnet etc. The main target of this survey is to give nearly full image of sentiment classification techniques. Survey paper provides the comprehensive overview of recent and past research on sentiment classification and provides excellent research queries and approaches for future aspects

An investigation of social media data during a product recall scandal

ABSTRACTAs social media has become an important part of modern daily life, users often share product opinions online and these tend to spike when large companies undergo crises. This paper investigates customer online responses to a large company crisis by uncovering hidden insights in social media comments and presents a framework for handling social media data and crisis management. Analysis of textual Facebook data from users responding to the 2013 horsemeat scandal is presented. In this study, we used a novel comprehensive data analysis framework alongside a text-mining framework to objectively classify and understand customer perceptions during this horsemeat scandal. This framework provides an effective approach for investigating customer perception during a company crisis and measures the effectiveness of crisis management practices which the company has adopted. Our analyses show that social media can provide important insights into customer behaviour during crisis communications.

Assessing the Feasibility of a Social Media to Promote Weight Management Engagement in Adolescents with Severe Obesity: Pilot Study.

BackgroundSevere obesity in adolescents has deleterious physical and psychological complications necessitating frequent multi-disciplinary clinic visits. Greater treatment engagement has been equated with weight-loss. However, traditional medical weight-loss programs for adolescents have high attrition rates. Social media is widely used by adolescents and may enhance medical weight management engagement and success.ObjectiveThe first objective was to examine the acceptability and feasibility of using a private social media group as an adjunct to medical weight management in youth ages 14 to 20 years with severe obesity [body mass index (BMI) ≥ 35 kg/m2]. The second objective was to pilot test the use of social media to improve treatment engagement and decrease attrition rates.MethodsIn this single arm, 12 week pre-post study, participants attended individual clinic visits and participated in a moderated private social media group that received nutrition, exercise, and behavior change social media communications or “posts” 3 to 4 times/week. Youth commented and/or liked posts from the moderator and each other. Social media engagement was measured with the number of likes and comments on social media. Clinic attrition was compared, measuring clinic visit attendance 12 weeks prior, during, and after the intervention with mixed linear regression models. Correlations of social media engagement with changes from baseline for BMI, BMI-z score, and psychosocial measures were fit.ResultsAll 13 enrolled youth completed the study and reported that the group was enjoyable, helpful, reinforced their weight management program, and would recommend using social media to support other youth. The pilot trial was acceptable and feasible. Youth mean weekly engagement (likes + comments) in social media was greater than once a day (8.6 ±3.6). Compared to 12 weeks prior to the intervention, there was no significant decrease in clinic visit attendance at the end of the intervention (M=.231, P=.69) or 12 weeks at the conclusion of the intervention (M=.589, P=.28). Increased social media comments correlated with weight change (r=–.633, P=.04).ConclusionsThis pilot trial demonstrated that the use of social media as an adjunct to medical weight management was feasible and acceptable to adolescents with severe obesity. Based upon these preliminary findings, social media may be an effective way to mitigate attrition from obesity treatment programs, and improve health outcomes in this high-risk population.

Virtual moderator using analysis of user trajectories on a semantic map based on convolutional neural networks

Comments in social media require moderation to avoid escalation of insults and overt aggression. Given the current large volume of communications, this moderation can only be done automatically. Modern state-of-the-art techniques that allow one to automatically detect insults in text have their limitations. For example, methods using forbidden words may not be efficient in online chats. The main drawback, however, is related to the fact that evaluation is done without taking into account user’s behavior in the past. The main purpose of the present work is to develop a virtual moderator that is free of this flaw. The approach is based on a set of convolutional neural networks and a semantic map, together taken as the basis of the model. The networks are trained for sentiment extraction from sentences found in available corpora. The virtual moderator matches coordinates of the 2-D semantic map to each fragment of the given text. One map axis represents offensiveness of the fragment, and the other the sentiment (positivity or negativity). The virtual moderator takes into account behavioral history of each user. However, newer messages determine the output to a higher extent compared to the older ones. The implemented prototype of a virtual moderator was tested on conversations in the Telegram app. Participants were instructed to communicate freely. At the same time, a confederate was introduced into the group, whose role was to play a bully, to insult others, and otherwise behave inappropriately. Afterwards, regular participants were asked to evaluate the correctness of Virtual Moderator’s decisions. The results show 73% accuracy of the Virtual Moderator.

Effects of emotion and topic area on topic shifts in social media discussions

Nowadays, discussing and commenting interactively on an article in Internet-based social media platforms is pervasive. The topic of a comment/reply in these discussions occasionally shifts, sometimes drastically and abruptly, other times slightly, away from the topic of the article. In this paper, we model and study the topic shift phenomena in article-originated social media comments, and identify quantitatively the effects on topic shifts of comments’ (1) emotion levels (of various emotion dimensions), (2) topic areas, and (3) the structure of the discussion tree. We then propose and evaluate a new approach to measure and visualize named emotion scores of comment sets. We show that, with a better understanding of the topic shift phenomena in comments, personalized automated systems can be built to cater to comment-browsing and comment-viewing needs of different users.

How the Chinese Government Fabricates Social Media Posts for Strategic Distraction, Not Engaged Argument

The Chinese government has long been suspected of hiring as many as 2 million people to surreptitiously insert huge numbers of pseudonymous and other deceptive writings into the stream of real social media posts, as if they were the genuine opinions of ordinary people. Many academics, and most journalists and activists, claim that these so-called 50c party posts vociferously argue for the government’s side in political and policy debates. As we show, this is also true of most posts openly accused on social media of being 50c. Yet almost no systematic empirical evidence exists for this claim or, more importantly, for the Chinese regime’s strategic objective in pursuing this activity. In the first large-scale empirical analysis of this operation, we show how to identify the secretive authors of these posts, the posts written by them, and their content. We estimate that the government fabricates and posts about 448 million social media comments a year. In contrast to prior claims, we show that the Chinese regime’s strategy is to avoid arguing with skeptics of the party and the government, and to not even discuss controversial issues. We show that the goal of this massive secretive operation is instead to distract the public and change the subject, as most of these posts involve cheerleading for China, the revolutionary history of the Communist Party, or other symbols of the regime. We discuss how these results fit with what is known about the Chinese censorship program and suggest how they may change our broader theoretical understanding of “common knowledge” and information control in authoritarian regimes.

Sprint’s Social Media Ninja Program: A Model for Teaching Consumer Relations

This study reviews the application of a new training model, Sprint’s Social Media Ninja program, an innovative approach to using new media to initiate change. Sprint recognized change management must occur from employee ambassadors to relevant audiences including consumers and other employees. By teaching volunteer employees the strategic message savvy and tactical strengths needed to address social media comments about Sprint, “Social Media Ninjas” have become active change agents in Sprint’s reputation management strategies, product launches, and turnaround story. These unmasked company employees volunteer to address questions, concerns, and comments about the company, as well as to start original conversations.

Health professionals’ reactions to patients’ social media posts about treatment

Introduction : Social media use for health purposes is gaining ground, and individuals and organizations are increasingly using social media to inform about and discuss health-related issues and even to offer treatment [1,2]. The health-related use of the Internet and the social media can empower patients in dealing with their own health and illness [3]. However, there has been concern relating to how social media may influence the health professional-patient relationship [4], and guidelines have been developed on health professionals’ social media use [5]. Methods : In a pilot study, we performed in-depth interviews with five healthcare providers (one doctor, two psychologists, and two nurses) working at a Norwegian psychiatric university clinic. They were asked about their work-related experiences with social media and especially whether they had experienced that their patients had discussed their treatment on social media. The interviews were analyzed drawing on the method of content analysis. Results : All had private social media accounts, but none had extensive work-related experience with social media. None had posted on social media as part of their work. Three had read social media comments posted by their patients relating to treatment. Patients’ posting of criticism was perceived of as potentially problematic for the health professional-patient relationship. Discussion, limitations, and lessons learnt : It may represent a challenge for health providers when patients discuss and evaluate their treatment on social media. While patients may post about their interaction with health professionals and treatment, confidentiality makes this off limits to health professionals [4,5]. Moreover, such posts can in some cases reflect ongoing problems in the health professional-patient relationship that should be addressed by the patient and the health professional. If health professionals feel that they are publicly unfairly evaluated by patients, this might provoke negative feelings towards patients and thereby negatively influence the health professional-patient relationship [3-6]. The study suggests that there is a need for discussion amongst health professionals regarding how to handle online criticism in a professional manner. This was a small exploratory study and more research is needed to confirm its findings. References : 1. Gabarron E, Serrano JA, Wynn R, Armayones M. Avatars using computer/smartphone mediated communication and social networking in prevention of sexually transmitted diseases among North-Norwegian youngsters. BMC Med Inform Decis Mak 2012;12:120. 2. Gabarron E, Serrano JA, Wynn R, Lau AY. Tweet content related to sexually transmitted diseases: no joking matter. J Med Internet Res 2014;16(10):e228. 3. Santana S, Lausen B, Bujnowska-Fedak M, Chronaki CE, Prokosch HU, Wynn R. Informed citizen and empowered citizen in health: results from an European survey. BMC Fam Pract 2011;12:20. 4. Moubarak G, Guiot A, Benhamou Y. Facebook activity of residents and fellows and its impact on the doctor-patient relationship. J Med Ethics 2011;37:101-4. 5. Mansfield SJ, Morrison SG, Stephens HO, et al. Social media and the medical profession. Med J Aust 2011;194:642-4. 6. Wynn R. Provider-patient interaction: A corpus-based study of doctor-patient and student-patient interaction, Oslo, Norway, Norwegian Academic Press, 1998.

DERIV: distributed brand perception tracking framework

Determining user’s perception of a brand in short periods of time has become crucial for business. Distilling brand perception directly from people’s comments in social media has promise. Current techniques for determining brand perception, such as surveys of handpicked users by mail, in person, phone or online, are time consuming and increasingly inadequate. The DERIV system distills storylines from open data representing direct consumer voice into a brand perception. The framework summarizes perception of a brand in comparison to peer brands with in-memory distributed algorithms utilizing supervised machine learning techniques. Experiments performed with open data and models built with storylines of known peer brands show the technique as highly scalable and accurate in capturing brand perception from vast amounts of social data compared to sentiment analysis.

#thisshowsucks! The Overpowering Influence of Negative Social Media Comments on Television Viewers

Can the comments of a few viewers on social media affect viewers’ perceptions of audience sentiment or their own program enjoyment? If so, do the effects of comments vary based on their valence or placement during programming? We conducted a 2 (positive vs. negative tweets) x 2 (beginning vs. end of program) factorial experiment with an additional control condition (N = 196) to answer these questions. Negative comments undermined perceived bandwagon support for the program and reduced enjoyment, regardless of contextual or trait moderators. Findings suggest that the effects of social television may be attributable to systematic processing of bandwagon cues.