Emotions present in social media content shape its diffusion. This study seeks to comprehensively examine the impact created by emotions of a social media message on its diffusion. Centered on a non-domain specific Twitter dataset, the authors define several measurement constructs to quantify the tweet diffusion process namely, speed, size, half-life, diffusion potential, and engagement. Since a message may express a single dominant emotion or multiple categories of emotions, the current study focuses to investigate the influence of emotions in the single label as well as multi-label setting. Through extensive statistical analyses (Multivariate Analysis of Variance and Regression), we find that the impact of emotions on diffusion constructs was statistically significant. The findings shed light on how emotions aid or hinder the spread of information through social media. Specifically, the tweets containing joy or contempt as primary emotion attained faster and stronger diffusion. In contrast, anger or fear as primary emotion in tweets contributed to slower and weaker diffusion. Also, the combination of one or more positive and negative emotions increased the diffusion outcome.Povzetek: Analiziran je vpliv pozitivnih in negativnih čustev na hitrost razširjanje čivka oz. tvita.

No abstract

The widespread adoption of social media applications in today's internet world has created a revolution in the field of Textual emotion mining (TEM). TEM has gained a significant amount of interest in the past few years where a large community of researchers is focusing on the efficient extraction of emotions, ignoring the influential behavior of emotions. This paper presents a unique study to focus on the outcome of research on Textual Emotion Mining instead on the process of emotion mining. It classifies the output of TEM by presenting an emotion-output model. It also discusses the influential capability of emotions in various domains, shedding light on various new avenues by analyzing the different work done in this area.

Textual Emotion Mining (TEM) tackles the problem of analyzing the text in terms of the emotions, it expresses or evokes. It focuses on a series of approaches, methods, and tools to help understand human emotions. The understanding would play a pivotal role in developing relevant systems to meet human needs. This work has drawn significant interest from researchers worldwide. This article carries out a science mapping analysis of TEM literature indexed in the Web of Science (WoS), to provide quantitative and qualitative insight into the TEM research. To explain the evolution of mainstream contents, various bibliometric indicators and metrics are used which identify annual publication counts, authorship patterns, performance of countries/regions, and institutes. To further supplement this study, various types of network analysis are also performed like co-citation analysis, co-occurrence analysis, bibliographic coupling, and co-authorship pattern analysis. Additionally, a fairly comprehensive manual analysis of top-cited and most-used journal and proceeding papers is also conducted to understand the growth and evolution of this domain. As per the authors' knowledge, this manuscript provides the first thorough investigation of TEM's research status through a bibliometric examination of scientific publications. Expedient results are recorded that will allow TEM researchers to uncover the growth pattern, seek collaborations, enhance the selection of research topics, and gain a holistic view of the aggregate progress in the domain. The presented facts and analysis of TEM will help the researchers' fraternity to carry out the future study.