تجزیه و تحلیل بحث توییتر در مورد انتخابات ریاست جمهوری سال ۲۰۱۶ اتریش
ترجمه نشده

تجزیه و تحلیل بحث توییتر در مورد انتخابات ریاست جمهوری سال ۲۰۱۶ اتریش

عنوان فارسی مقاله: سیاست، گرایشات و اطلاعات نادرست: تجزیه و تحلیل بحث توییتر در مورد انتخابات ریاست جمهوری سال ۲۰۱۶ اتریش
عنوان انگلیسی مقاله: Politics, sentiments, and misinformation: An analysis of the Twitter discussion on the 2016 Austrian Presidential Elections
مجله/کنفرانس: رسانه ها و شبکه های اجتماعی آنلاین - Online Social Networks and Media
رشته های تحصیلی مرتبط: مهندسی فناوری اطلاعات
گرایش های تحصیلی مرتبط: اینترنت و شبکه های گسترده، مدیریت سیستم های اطلاعاتی
کلمات کلیدی فارسی: مطالعه موردی، تجزیه و تحلیل شبکه، مبارزات انتخاباتی سیاسی، تحلیل احساسات، توییتر
کلمات کلیدی انگلیسی: Case study، Network analysis، Political campaigning، Sentiment analysis، Twitter
نوع نگارش مقاله: مقاله پژوهشی (Research Article)
شناسه دیجیتال (DOI): https://doi.org/10.1016/j.osnem.2017.12.002
دانشگاه: Vienna University of Economics and Business WU Vienna, Austria
صفحات مقاله انگلیسی: 14
ناشر: الزویر - Elsevier
نوع ارائه مقاله: ژورنال
نوع مقاله: ISI
سال انتشار مقاله: 2018
شناسه ISSN: 2468-6964
فرمت مقاله انگلیسی: PDF
وضعیت ترجمه: ترجمه نشده است
قیمت مقاله انگلیسی: رایگان
آیا این مقاله بیس است: بله
کد محصول: E11228
فهرست مطالب (انگلیسی)

Abstract

1- Introduction

2- Event of study

3- Research questions and approach synopsis

4- Data analysis

5- Discussion

6- Related work

7- Conclusion

References

بخشی از مقاله (انگلیسی)

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.