1- Introduction

2- Related works

3- Methodology

4- Dataset statistics and preprocessing

5- Feature sets

6- Experimental results and insights

7- Conclusion

References

Abstract

The last decade witnessed an enormous growth in popularity of several social media platforms. Although these platforms are generally meant to share information and opinions, their ubiquity is being increasingly exploited for spreading news and events in real time. Hence these platforms have become a natural choice for news agencies for getting updates, comments and experts’ opinion of ongoing events which is crucial to understand the societal impact and for writing reports/editorial. However, with the plethora of these platforms available, each having its own uniqueness in content presentation, spreading patterns and also in the user interests, a comparative study of the efficacy of these platforms for different journalistic purposes would be useful. In this paper, we perform a comparative study of two leading social media platforms Reddit and Twitter. We have analyzed Reddit comments and Twitter feeds of six news categories to establish the efficacy of these platforms in terms of different journalistic requirements. Observations reveal that there exist significant differences across these platforms that can be suitably exploited depending upon the scope of the requirements; for example, while Twitter is a better choice for the evolutionary study of events, Reddit is the more natural choice for exploration during the initial phase of any event. While the availability and spread of updated information on Twitter can be key in emergency and disaster situations, critical analysis of posts in Reddit can be important for editorials.

Introduction

Increasing use of smartphones and high penetration of data connectivity have lent to the enormous growth in popularity of social media platforms. The propensity of users to share news updates and opinions over social media platforms makes them an important source for obtaining news and reactions in real time. According to Pew Research center survey report in 20171, 67% of the adults get news from social media. The news agencies consider these platforms as a rich source of information and use them for mining news and user opinions [1]. The recently released Cision 2017 Global Social Journalism revealed that nearly half of the journalists can not do their work without help of social media2. Journalists use the social media for meeting various journalistic requirements that include obtaining breaking news and real-time news-updates, finding opinions, enriching arguments, confirmation of facts and deriving information source [2–5]. A large variety of social media platforms are currently in place that includes, microblogs, discussion forums, question-answer sites and social news aggregators to name a few [6]. Although most of these platforms are used for sharing facts and updates, sharing users’ view and for the propagation of news events, the microblog and the news aggregator sites are currently playing a predominant role in event identification, news propagation, and opinion sharing [7–9]. Twitter has traditionally been recognized as the major microblogging platform for obtaining such updates, however, issues like short content lengths, large vocabulary gaps and the inherently noisy nature of the tweets pose subtle challenges in mining the required information [10,11]. On the other hand with the increasing popularity of news aggregator sites like Reddit, journalists are exploring these platforms for meeting the journalistic requirements mentioned above. Thus, considering the diverse features, functionalities and user behavior of the microblog and news aggregation platforms, one of the current needs is to determine the suitability of a platform in mining relevant information depending upon the specificity of the requirement.

Monitoring the emergence of emerging technologies helps managers and decision makers to identify development trends in emerging technologies is crucial for government research and development (R&D), strategic planning, social investment, and enterprise practices. Researchers usually use academic papers and patent data to identify and monitoring the trends of emerging technologies from a technological perspective, but they rarely make use of social media data (e.g., such as Twitter data) related to emerging technologies. Analysis of this social media data is of great significance to understand the emergence of emerging technologies and gain insight into development trends. Therefore, this paper proposes a framework that uses patent analysis and Twitter data mining to monitoring the emergence of emerging technologies and identify changing trends of these emerging technologies. The perovskite solar cell technology is selected as a case study. In this case, we used patent analysis to monitoring the evolutionary path of perovskite solar cell technology. We applied Twitter data mining to analyze Twitter users’ sense of, response to, and expectations for this perovskite solar cell technology. We also identified the professional types of Twitter users and examined changes in their topics of interest over time to track the emergence of perovskite solar cell technology. We analyzed a comparison of the results of patent analysis and Twitter data mining to identify development trends of perovskite solar cell technology. This paper contributes to our understanding of how technologies emerge and develop, as well as the technology forecasting and foresight methodology, and will be of interest to solar photovoltaic technology R&D experts.

Introduction

In recent years, emerging technologies with disruptive characteristics have been rapidly emerging. Major technological advancements include information technology, nanotechnology, biotechnology, and new material technology. The emergence and development of these emerging technologies not only has changed existing industries but also has created new industries that have a significant impact on the socioeconomic structure (Day et al., 2000). Identifying future changing trends of these emerging technologies as early as possible is crucial for government and enterprise research and development (R&D) strategic planning to gain a first-mover advantage in the competitive business environment. Many managers and decision makers are aware of the significance of understanding of the emergence path, and identifying the future development trends of these emerging technologies for their organization’s competitiveness and sustainable development (Li et al., 2015). This strategic concern issue raises one question: Can we identify the future development trends of these emerging technologies given a better understanding of how they emerge? In response to this question, this paper develops a framework to monitoring the emergence of emerging technologies and identify future changes trends of these emerging technologies by using patent analysis and Twitter data mining.

Purpose – Social media have recently become an important strategic marketing tool to increase firm value. Based on an integrated theoretical framework, this study aims to examine the market reaction at the time of the creation of a Twitter platform for 312 firms from the Fortune 500 firms. Design/methodology/approach – To test the hypotheses related to the effect of social media platforms on firm value, the event history analysis (EHA) was used, also known as event study, usually designed to examine the impact of a historical phenomenon for the US Fortune 500 firms that developed a Twitter platform. Findings – A significant market reaction was found around the starting date of Twitter activities for the subsample of firms that are not contaminated by any other corporate announcements, but not for the overall sample. The market reaction is higher for firms with two-way interaction strategies rather than one-way messaging in both the uncontaminated subsample and the overall sample. It is higher in smaller firms, firms with losses and those with a family and/or a dominant shareholder. Further, firms in the contaminated subsample are likely to follow a two-way strategy after a positive revision of their earnings per share. We have run several robustness checks, including cross-validation on a holdout sample, and these findings remain consistent. Research limitations/implications – The integrated theoretical framework is another significant contribution. To our knowledge, this is the first study across disciplines that integrates the social exchange theory (SET), social representation theory (SRT), social network analysis (SNA), social identity theory (SIT), signaling theory (ST) and the impression management theory (IMT) into one framework that is built around information as a resource and social interaction. Practical implications – The results suggest that Twitter can be used to add value if firms interact and reciprocate with the various stakeholders. Social implications – Firms using social media must interact and reciprocate with the various stakeholders. Originality/value – This research is different than the published research on this topic in that it examines the impact on stock prices of the introduction of a specific social media platform, i.e. Twitter. The present results of the paper add to the prior research on database marketing and show that marketing “with” the customer is adding more value than marketing “to” the customer. The use of the net extends the scope of database marketing into a certain form of interaction marketing with “face-to-face” interaction within the relationships between the firm and its customers. Finally, the conditions under which social media platforms are used in an interactive manner are shown, and depicts that firms are more likely to use a two-way interactive strategy following a one-year period of positive momentum.

Introduction

One bad Twitter “tweet” can lose companies as many as 30 customers (Sarah Shannon – November 25, 2009, Bloomberg). Social media is a technological phenomenon that facilitates communication and collaboration embedded in humans’ lives (Goldenberg et al., 2012; Tucker, 2014). Firms adopt policies and plans to achieve business goals in which insights gleaned from information exchange support the decision-making process (Bayus, 2013; Holsapple and Singh, 2000). Social media platforms are more frequently used as strategic marketing tools and channels to promote new products – goods and services (Park et al., 2010), to reach, observe and get closer to customers (Gopinath et al., 2013), to better understand their individual preferences (Li and Shiu, 2012) and to build customer equity (Drèze and Bonfrer, 2008). Further, these platforms provide stakeholders with several benefits and services including location-based recommendations (Zhao and Lu, 2012), user reviews (Hoehle et al., 2012) and development of personal and company brands, which is likely to result in a greater incremental value. However, the credibility of a social media strategy depends on how effective it is in helping the firm achieve its strategic goals and objectives. Despite the increasing number of companies using social media platforms, little is known about how the use of a particular social medium may affect the value of originator firms. In a recent survey looking at the use of social media by trading and investment firms, OneMarketData (2013) – a leader in tick data management and analytics – indicated a growing interest in social media in financial markets, but also some skepticism about the ability of social media to transmit quality signals to the market[1].

1- Introduction

2- Event of study

3- Research questions and approach synopsis

4- Data analysis

5- Discussion

6- Related work

7- Conclusion

References

Abstract

In this paper, we provide a sentiment analysis of the Twitter discussion on the 2016 Austrian presidential elections. In particular, we extracted and analyzed a data-set consisting of 343645 Twitter messages related to the 2016 Austrian presidential elections. Our analysis combines methods from network science and sentiment analysis. Among other things, we found that: a) the winner of the election (Alexander Van der Bellen) predominantly sent tweets resulting in neutral sentiment scores, while his opponent (Norbert Hofer) preferred emotional messages (i.e. tweets resulting in positive or negative sentiment scores), b) negative information about both candidates continued spreading for a longer time compared to neutral and positive information, c) there was a clear polarization in terms of the sentiments spread by Twitter followers of the two presidential candidates, d) the winner of the election received considerably more likes and retweets, while his opponent received more replies, e) the Twitter followers of the winner substantially participated in the spread of misinformation about him.

Introduction

In recent years, social media have become an important channel for politicians to address the public, making them more accessible to their prospective voters [1–3]. Although social media are often used to disseminate informative content, such as event announcements on a candidate’s public appearances, recent studies have shown that social media are also used for spreading misinformation as a part of political propaganda [4–6]. In this context, the emotional dimension of a social media discussion [7] is of particular importance as an emotional debate over a controversial topic often develops more dynamically and unpredictably than an objective discussion. Sentiment analysis methods [8] help classify and understand the users’ feelings about a topic of interest. However, the sheer complexity of socio-technical systems [9,10] and the big data characteristics of complex networks [11,12] make the analysis of social media events a difficult task [13,14]. In this context, case studies of real-world political campaigns are of particular interest because they help understand human behavior, detect patterns, and identify generic approaches for analyzing user behavior in online social networks (see, e.g., [2,15–19]). In this paper, we provide a comprehensive sentiment analysis of the Twitter discussion related to the 2016 Austrian presidential elections and show that during political campaigns conveying emotional content is not always advantageous for the respective political candidate. In particular, we extracted and analyzed a data-set consisting of 343,645 Twitter messages. The resulting data-set is multi-dimensional, including temporal data, structural data (such as the corresponding topic/hashtag network), as well information on the user’s emotions that are expressed in the content of the messages. In addition to sentiment polarities, our analysis also identifies specific emotions about each candidate that are conveyed in tweets posted by other Twitter users. The remainder of this paper is structured as follows. First, we give an overview of the election event in Section 2. Next, Section 3 provides an approach synopsis and discusses the guiding research questions for our study. Subsequently, Section 4 presents our sentiment analysis of the Twitter discussion on the 2016 Austrian presidential elections. In Section 5, we further discuss our findings as well as the limitations of our study. Section 6 discusses related work and Section 7 concludes the paper.

1- Introduction

2- Literature Review

3- Data Collection Through Web Scraping

4- Model Creation for Labeling

5- Analysis of User Profiles and Tweets

6- Conclusion

References

Abstract

This project is focused on exploring the appealing trends in the profiles of different users, with respect to different aspects of the user demographics which includes gender, whether the user is an individual or an organization, psychological and their academic background, who had discussed psychological articles on Twitter. To perform this task we retrieved the details of psychological articles, related tweets and twitter user details by web scraping using python. Then to assign suitable labels, we have used the Rapid Miner to create a model using training data. After the labeling process, we have analyzed the patterns in user profiles. Further, the tweet contents were analyzed focusing on the psychological topics more discussed by the users. This analysis will give the insight to psychological researchers, what topics are being discussed by the Twitter users and what are trending topics among Academic communities and Nonacademic Communities.

Introduction

The introduction of social media is one of the significant events in this modern era to create a huge impact on our lives. These social networks are very widespread in this decade since this internet enabled services such as smart phones, tablets, and personal computers. Nowadays users use social media to share information and thoughts with each other. This paves way for a large amount of data to be shared on social media. In short communication content such as tweets was important to study as it was believed to have an effect on the society. According to Riff, Lacy, and Fico, social affiliations such as family and community involvement were important predictors of peoples attitudes and behaviors, and a network of personal influence was identified as key factors influencing their decisions.[1, 2] In the academic community, some individuals use these tweets as a medium to announce their publications of new articles and to cite those articles in the tweets. Due to the big data associated with these social media, its not possible for us to apply similar approaches used in traditional ways available in research evaluation to analyze about the impacts of research articles in the society. Hence, tools such as Altmetrics play an important role in analyzing research impact based on social media activities.[3] Altmetrics measures the impact of the social and academic use of research in different fields. It includes academic comments, citation and reuse and social media components such as tweets, shares, as well as ratings. Further, this is defined broadly and offers a variety of avenues to get a complete idea of measuring the research impact. The metrics of research aim to quantify the significance of research such as how many highly cited papers an author has produced, how many publications of articles in heavily cited journals and how many times the article has been cited. Metrics are an attempt to provide researchers a direct response to the value of work. Psychology field is selected for this project as this field was having a higher citation rate and one of the top Altmetrics coverage when compared with other disciplines.

Customer engagement is drastically improved through Web 2.0 technologies, especially social media platforms like Twitter. These platforms are often used by organizations for marketing, of which creation of numerous spam profiles for content promotion is common. The present paper proposes a hybrid approach for identifying the spam profiles by combining social media analytics and bio inspired computing. It adopts a modified K-Means integrated Levy flight Firefly Algorithm (LFA) with chaotic maps as an extension to Firefly Algorithm (FA) for spam detection in Twitter marketing. A total of 18,44,701 tweets have been analyzed from 14,235 Twitter profiles on 13 statistically significant factors derived from social media analytics. A Fuzzy C-Means Clustering approach is further used to identify the overlapping users in two clusters of spammers and non-spammers. Six variants of K-Means integrated FA including chaotic maps and levy flights are tested. The findings indicate that FA with chaos for tuning attractiveness coefficient using Gauss Map converges to a working solution the fastest. Further, LFA with chaos for tuning the absorption coefficient using sinusoidal map outperforms the rest of the approaches in terms of accuracy.

Introduction

Use of social media platforms like Twitter, Facebook, YouTube, Instagram and LinkedIn etc. is indispensable for information diffusion in the current world (Kaplan and Haenlein 2010; Lenhart et al. 2010; Mui and Whoriskey 2010). The platforms are being used by the people for communicating their opinions, experiences and ideas and are faster than traditional media in diffusion of information (Bakshy et al. 2012). Following individual users, organizations and businesses are also present in the platforms to connect with different stakeholders. The profiles of businesses are predominantly for content promotion and knowledge dissemination surrounding products and services, making the social media platforms the backbone of digital marketing. The social media presence is utilized to promote the content to a larger audience (Hanna et al. 2011). Needless to say that platforms are now flooded with information from various sources by multiple organizations competing for the users’ attention (Romero et al. 2011). Among the social platforms, Twitter is one of the fasting growing micro blogging platforms that helps people communicate and share their opinion using short messages (Huberman et al. 2008). The extant knowledge highlights that about 54% of the Fortune 50 organizations have Twitter account and 37% of these have multiple profiles. Further, 85% of these organizations use Twitter for news dissemination (Case and King 2011). When compared among social media platforms, Twitter has the highest usage for business (78%), followed by LinkedIn (74%), and Facebook (44%) (Go and You 2016). These high usage statistics of Twitter by businesses and organizations pose greater risk of spams. Twitter by itself uses verified accounts for profiles of various celebrities, political figures and organizations but it is only 0.061% of all Twitter accounts.