Machine learning for predicting elections in Latin America based on social media engagement and polls

The popularity of contemporary social media (SM) has impacted democratic practices, and the success of presidential campaigns is frequently attributed to SM performance. Within this new scenario, many methodological proposals that use SM data have been put forward for predicting election results. However, the most common approach, based on the volume and sentiment analysis of mentions on Twitter, has been frequently criticized and challenged. Thus, recent surveys have indicated new directions, such as the use of data from more than one SM platform, the adoption of nonlinear machine learning (ML) models, and the validation of methodologies and experiments in different elections. In this context, the present paper proposes SoMEN, the Social Media framework for Election Nowcasting, a framework composed of a process and an ML model for nowcasting election results based on SM performance as features and with offline polls as labeled data. It also defines SoMEN-DC, an execution strategy for SoMEN that enables continuous prediction during the campaign (DC). The proposed metrics and framework were applied to some of the main recent presidential elections in Latin America: Argentina (2019), Brazil (2018), Colombia (2018), and Mexico (2018). More than 65,000 posts were collected from the SM profiles of candidates on Facebook, Twitter, and Instagram with data from 195 presidential polls. Results have demonstrated that it was possible to achieve a high level of accuracy in predicting the final vote share of the candidates and to make daily predictions, providing competitive or better results than the traditional polls. The strategies put forward in this study have attempted to address several of the current challenges in this research area and have indicated a new manner of how to face the problems. They may also be directly used for predicting future elections in similar scenarios.     

Examining government cross-platform engagement in social media: Instagram vs Twitter and the big lift project

As governments are increasingly turning to social media as a means of engaging the public, questions remain as to how they are actually using various social media platforms. Do specific platforms engender specific types of messages? If so, what are they, and how do they affect civic engagement, co-participation, and address citizen concerns? In this paper, we compare the use of Instagram and Twitter by ‘The Big Lift’, a bridge re-decking project completed by Halifax Harbour Bridges. Based on a content analysis of Instagram (n = 248) and Twitter (n = 1278) public posts, we found that Instagram was used as a more ‘informal’ narrative platform that promoted a clicktivist type of responses from the audience, whereas Twitter was a more ‘formal’ news platform that supported greater two-way communication between the organization and the audience. We conclude that by building and maintaining their active presence and following base on social media, and especially on Twitter, organizations can develop a capacity to address social concerns during disruptive events or infrastructure projects like ‘The Big Lift’.     

An artist ranking system based on social media mining

Currently users on social media post their opinion and feelings about almost everything. This online behavior has led to numerous applications where social media data are used to measure public opinion in a similar way as a poll or a survey. In this paper, we will present an application of social media mining for the art market. To the best of our knowledge, this will be the first attempt to mine social media to extract quantitative and qualitative data for the art market. Although there are previous works on analyzing and predicting other markets, these methodologies cannot be applied directly to the art market. In our proposed methodology, artists will be treated as brands. That is, we will mine Twitter posts that mention specific artists’ names and attempt to rank artists in a similar manner as brand equity and awareness would be measured. The particularities of the art market are considered mainly in the construction of a topic-specific user network where user expertise and influence is evaluated and later used to rank artists. The proposed ranking system is evaluated against two other available systems to identify the advantages it can offer.     

Flagging fake news on social media: An experimental study of media consumers' identification of fake news

Policymakers are taking action to protect their citizens and democratic systems from online misinformation. However, media consumers usually have a hard time differentiating misinformation from authentic information. There are two explanations for this difficulty, namely lazy reasoning and motivated reasoning. While lazy reasoning suggests that people may feel reluctant to conduct critical reasoning when consuming online information, the motivated reasoning theory points out that individuals are also thinking in alignment with their identities and established viewpoints. A proposed approach to address this issue is adding fact-checking flags in the hope that flags could alert people to information falsehoods and stimulate critical thinking. This study examines the impact of fact-checking flags on media consumers' identification of fake news. Conducting an experiment (n = 717) on Amazon Mechanical Turk, the study finds that experimental participants with different political backgrounds depend heavily on flag-checking results provided by flags. Flags are powerful to influence people's judgments in a way that participants have blind beliefs in flags even if the flag assessments are inaccurate. Furthermore, the study's results indicate that flag assessments made by professional fact-checkers or crowdsourcing are equally influential in shaping participants' identification. These observations provide public and private leaders with suggestions that fact-checking flags can significantly affect media consumers' identification of fake news. However, flags appear to have little ability to promote critical thinking in this experiment.     

Leveraging social media to achieve a community policing agenda

This research investigates the communication behavior and engagement strategies in the bilateral use of social media between law enforcement agencies and the communities they serve. It advances existing work by studying municipal level government actors in a new communications environment where social media now play an important part. Grounded in agenda setting theory, our analysis identifies police departments' social media issue priorities, analyzes the responses of their audiences to those communications, and directly compares followers' own conversation priorities with the police agenda. Our data set includes all the content from the Facebook and Twitter accounts of five similarly sized and demographically situated police departments in the U.S., plus all the tweets and posts from the followers or friends responding to those accounts over a 90-day period. We performed both manual coding and machine cluster analysis to elicit major threads of conversation. In addition to the data analytics, we conducted interviews with the five police departments to understand the similarities and differences in agenda priorities resulting from their social media goals and use.The study shows the priorities that comprise the police agenda, identifies both similarities and differences in what their audiences communicate among themselves about most frequently in the public safety domain relative to the police agenda, and finds evidence of positive response from the public to some of the agenda priorities communicated by the police. Our data also reveal that police are using social media interactively, which could, over time, advance community policing goals. The paper concludes by considering the implications of these findings for law enforcement and community policing and suggests directions for future research on agenda setting in this new media environment.     

Reinforcing attitudes in a gatewatching news era: Individual-level antecedents to sharing fact-checks on social media

Despite the prevalence of fact-checking, little is known about who posts fact-checks online. Based upon a content analysis of Facebook and Twitter digital trace data and a linked online survey (N?=?783), this study reveals that sharing fact-checks in political conversations on social media is linked to age, ideology, and political behaviors. Moreover, an individual’s need for orientation (NFO) is an even stronger predictor of sharing a fact-check than ideological intensity or relevance, alone, and also influences the type of fact-check format (with or without a rating scale) that is shared. Finally, participants generally shared fact-checks to reinforce their existing attitudes. Consequently, concerns over the effects of fact-checking should move beyond a limited-effects approach (e.g., changing attitudes) to also include reinforcing accurate beliefs.     

Leveraging library trust to combat misinformation on social media

One reason librarians are confident they have a role to play in fighting misinformation is the level of trust in libraries as institutions. Exactly how they might leverage that trust remains unclear and untested. Building on recent work in correcting health misperceptions on social media, this study tests whether libraries can leverage trust to combat misinformation online. Using a misperception about the influenza vaccine as a test case, an experiment (n?=?625) was conducted in fall 2018 using Amazon's Mechanical Turk. Results suggest that the misperception can be reduced, but not by library institutions. An unsuccessful follow-up (n?=?600) suggests that the effectiveness of the correction is season dependent and opens the possibility that libraries may yet play a role, but not necessarily because they are trusted. Future library proposals for combating misinformation need to be developed and tested within a broader contemporary misinformation research program.     

An index to publisher quality revisited: a partial replication

A replication of the Calhoun/Bracken 1983 study of publishers listed in Choice's annual Outstanding Academic Books was conducted. Data were gathered for the years 1988–1992 and compared with the results of the Calhoun/Bracken analysis. The results show that several of the commercial and university publishers that had been prominent in 1983 had been replaced by new presses. During the decade between studies, university publishers have showed marked growth as a percentage of the top 60 presses identified in the two studies. Collection development librarians need to be aware of the fluctuations in publisher output and reputations over time in the Outstanding Academic Book lists.     

An index to medical book reviews: a computer experiment

The experimental production of an index to book reviews of medical publications is described. Details of compilation and preparation of data for processing by an IBM 1440 computer are outlined. Results of a survey testing the response to the index in its present form by medical libraries are presented.     

Measuring the performances of politicians on social media and the correlation with major Latin American election results

Contemporary social media (SM) has strongly impacted democratic practices. The success of presidential campaigns is frequently attributed to being highly correlated with the candidates' social media performance, but there is no well-established method to measure this performance. Thus, this study aims to improve the understanding of a politician's performance on SM and its correlation with electoral results. Applying a new, recently-defined set of metrics, based on Zajonc's exposition theory and considering the interactions of users on politicians' profiles in multiple SM platforms, this research identifies statistical correlations between SM performance and the votes received in multiple elections. As case studies, this paper focuses on the most recent presidential elections in the four most populous countries in Latin America: Argentina (2019), Brazil (2018), Colombia (2018), and Mexico (2018). Data from more than 65,000 posts were collected from the SM profiles of the main candidates on Facebook, Twitter, and Instagram, starting from 300 days before the election days, and correlations with electoral results were calculated. The results demonstrated strong correlations between the defined metrics and the votes received, particularly the engagement per post, although there were differences among countries. On the other hand, we observed that there is zero or negative correlation between the number of posts and the electoral results.     

Consumer health information seeking in social media: a literature review

Objective

The objective of this literature review was to summarise current research regarding how consumers seek health‐related information from social media. Primarily, we hope to reveal characteristics of existing studies investigating the health topics that consumers have discussed in social media, ascertaining the roles social media have played in consumers’ information‐seeking processes and discussing the potential benefits and concerns of accessing consumer health information in social media.

Methods

The Web of Science Core Collection database was searched for existing literature on consumer health information seeking in social media. The search returned 214 articles, of which 21 met the eligibility criteria following review of full‐text documents.

Conclusion

Between 2011 and 2016, twenty‐one studies published explored various topics related to consumer information seeking in social media. These ranged from online discussions on specific diseases (e.g. diabetes) to public health concerns (e.g. pesticide residues). Consumers’ information needs vary depending on the health issues of interest. Benefits of health seeking on social media, in addition to filling a need for health information, include the social and emotional support health consumers gain from peer‐to‐peer interactions. These benefits, however, are tempered by concerns of information quality and authority and lead to decreased consumer engagement.     

Weibo diplomacy: Foreign embassies communicating on Chinese social media

This study investigates public diplomacy on a Chinese social media platform – Weibo. Examining the traffic (i.e., the number of followers) of 30 foreign embassies on Weibo, this study confirms that under certain conditions, digital diplomacy can be effective even in closed information systems. The results suggest that the number of followers an embassy has does not always increase with the economic size of a country or its bilateral economic relationships with China; instead, the number of followers of an embassy's Weibo account grows with the embassy's daily posts. Weibo allows embassies to unshackle the limitations imposed by their economic weight or geopolitical influence and to attract a larger audience through active daily posts. Using Weibo can also help foreign government representations in China manage changes in international relations. This study provides further insights into how social media have changed the traditional pattern of information flow not only in an open Internet environment but also in a closed social media environment where the Internet is under heavy censorship. It also proposes some recommendations for policy makers.     

Public library needs assessment to build a community-based library: Triangulation method with a social media data analysis

Public libraries must respond to the needs of the communities they serve in order to remain relevant, but assessing these needs is especially challenging in the midst of the rapid development of information technology. This study examines needs assessments to understand the user community, library services, and expected sources to determine user needs regarding space and services. The research employed a mixed-method approach including semi-structured interviews, a questionnaire assessment, and Social Network Site (SNS) big data analysis. The study assessed the needs and characteristics of users and non-users at Yongsan-gu Public Library, South Korea. Data collected were used to examine how the library was affected by COVID-19, the steps it taken to adjust and provide services, and how users have adapted to library use during the pandemic. The research results provide direction for building a future public library in regions that lack cultural infrastructure. The results also demonstrate that it is necessary to construct infrastructure linked to cultural projects by creating complex cultural and user-oriented spaces.     

Predicting elections from social media: a three-country,three-method comparative study

ABSTRACT

This study introduces and evaluates the robustness of different volumetric, sentiment, and social network approaches to predict the elections in three Asian countries – Malaysia, India, and Pakistan from Twitter posts. We find that predictive power of social media performs well for India and Pakistan but is not effective for Malaysia. Overall, we find that it is useful to consider the recency of Twitter posts while using it to predict a real outcome, such as an election result. Sentiment information mined using machine learning models was the most accurate predictor of election outcomes. Social network information is stable despite sudden surges in political discussions, for e.g. around elections-related news events. Methods combining sentiment and volume information, or sentiment and social network information, are effective at predicting smaller vote shares, for e.g. vote shares in the case of independent candidates and regional parties. We conclude with a detailed discussion on the caveats of social media analysis for predicting real-world outcomes and recommendations for future work.     

From e-government to we-government: Defining a typology for citizen coproduction in the age of social media

This paper examines the evolution of citizen coproduction in the age of social media, web 2.0 interactivity, and ubiquitous connectivity. The paper first discusses the re-emergence of citizen coproduction – whereby citizens perform the role of partner rather than customer in the delivery of public services – as a fashionable policy option in the face of persistent budget deficits and the advent of new channels for mass collaboration. Finding a plethora of competing labels, models, and concepts for coproduction in the age of social media, the paper proposes a unified typology to support systematic analysis based on the overarching categories of “Citizen Sourcing,” “Government as a Platform,” and “Do-It-Yourself Government.” To demonstrate its use, the typology is applied to leading U.S. government implementations. The paper concludes with a discussion of the potential implications for public administration, the remaining limitations and rising social concerns, and the possible emergence of a new social contract that empowers the public to play a far more active role in the functioning of their government.     

Twitter as a tool for citizen engagement: An empirical study of the Andalusian municipalities

This paper provides a general overview of the way local governments use Twitter as a communication tool to engage with their citizens. More concretely, it tries to identify factors associated with both the channel activity and citizen engagement, to understand the relationship between media type and citizen engagement and to analyse whether different content generated different levels of engagement. A sample of the 29 most populated Andalusian local governments is examined. The results show that the majority of Andalusian local governments have an official corporate Twitter account with certain level of activity. There is no, however, a significant relationship between the population of a municipality and its citizen's engagement, and there is a significant negative relationship between audience and engagement and between activity and engagement. The findings of the study also show the particular media and content types generate higher engagement than others. This paper contributes to the literature on social media and has practical implications for local governments.     

Using ICTs to create a culture of transparency: E-government and social media as openness and anti-corruption tools for societies

In recent years, many governments have worked to increase openness and transparency in their actions. Information and communication technologies (ICTs) are seen by many as a cost-effective and convenient means to promote openness and transparency and to reduce corruption. E-government, in particular, has been used in many prominent, comprehensive transparency efforts in a number of nations. While some of these individual efforts have received considerable attention, the issue of whether these ICT-enabled efforts have the potential to create a substantive social change in attitudes toward transparency has not been widely considered. This paper explores the potential impacts of information and ICTs – especially e-government and social media – on cultural attitudes about transparency.     

Trust,norms and networks in social media environmental mobilization: a social capital analysis of Under the Dome in China

After the Spring Festival of 2015, an environmental documentary called Under the Dome in which Chai Jing acted as a narrator has caused the public opinion to boom in social media in China. Why could a documentary released by internet attract so much attention and spark such a heated discussion? What are the factors that have influence on the information dissemination of social media environmental mobilization? This paper employed case study and semi-structure interviews to reveal these questions. It turned out that individual trust, group norms and information networks of the social capital played an important role in the social media environmental mobilization. The reason why Under the Dome became a hot issue was that it made full use of the social media users’ trust on the mobilizer, aroused the mobilized’s emotions and combined the network channels of social relationships.