Good Friends, Bad News - Affect and Virality in Twitter (1101.0510v1)

Abstract: The link between affect, defined as the capacity for sentimental arousal on the part of a message, and virality, defined as the probability that it be sent along, is of significant theoretical and practical importance, e.g. for viral marketing. A quantitative study of emailing of articles from the NY Times finds a strong link between positive affect and virality, and, based on psychological theories it is concluded that this relation is universally valid. The conclusion appears to be in contrast with classic theory of diffusion in news media emphasizing negative affect as promoting propagation. In this paper we explore the apparent paradox in a quantitative analysis of information diffusion on Twitter. Twitter is interesting in this context as it has been shown to present both the characteristics social and news media. The basic measure of virality in Twitter is the probability of retweet. Twitter is different from email in that retweeting does not depend on pre-existing social relations, but often occur among strangers, thus in this respect Twitter may be more similar to traditional news media. We therefore hypothesize that negative news content is more likely to be retweeted, while for non-news tweets positive sentiments support virality. To test the hypothesis we analyze three corpora: A complete sample of tweets about the COP15 climate summit, a random sample of tweets, and a general text corpus including news. The latter allows us to train a classifier that can distinguish tweets that carry news and non-news information. We present evidence that negative sentiment enhances virality in the news segment, but not in the non-news segment. We conclude that the relation between affect and virality is more complex than expected based on the findings of Berger and Milkman (2010), in short 'if you want to be cited: Sweet talk your friends or serve bad news to the public'.

Analysis of Affect and Virality on Twitter

The paper "Good Friends, Bad News: Affect and Virality in Twitter" by Hansen et al. presents a comprehensive investigation into the relationship between sentiment (affect) and the viral nature of content on Twitter. Addressing a gap between psychological theories and classical news diffusion theories, this paper empirically explores how affect influences the likelihood of content propagation via retweets.

Overview and Methodology

The researchers aim to test the hypothesis that while negative sentiment enhances the virality of news content on Twitter, positive sentiment sustains virality in non-news or social contexts. Three corpora were analyzed: a complete corpus of tweets regarding the COP15 climate conference, a random sample of tweets, and a general text corpus labeled with news attributes.

A Naive Bayes classifier was employed to categorize tweets as news or non-news based on the Brown Corpus, a benchmark dataset with a known distribution of news categories. Sentiment analysis was performed using a manually curated lexicon to assign affective scores to tweets. The paper constructed generalized linear models to evaluate the impact of sentiment on the probability of retweets.

Findings

1. Prevalence of News on Twitter:
   • The analysis revealed divergent results: approximately 23% of tweets in the random sample were classified as newsworthy, while the COP15 corpus demonstrated a higher occurrence (31%). This aligns with the hypothesis that Twitter, when focused on influential news events, becomes a significant channel for news dissemination.
2. Sentiment’s Role in Propagation:
   • In the context of general random tweets, the research demonstrated that negative sentiment did not enhance retweet likelihood. Contrarily, in the news context (as seen in the COP15 tweets), negative sentiment significantly fostered propagation, aligning with classical theories suggesting that negative content attracts more attention.
3. Interaction of Sentiment and News Classification:
   • The paper discovered that the interactive effect of negativity and news classification constructed stronger predictive models for virality. Negative news was retweeted more frequently than neutral or positive news. For non-news content, positive sentiment was more effective in driving engagement, corroborating with psychological frameworks on social sharing.

Implications and Future Perspectives

This paper provides crucial insights into the differential roles of affect in digital communication and challenges the universality of findings from previous studies focused on other platforms such as email. Importantly, it underscores the complexity of viral mechanisms on Twitter, which encompasses both traditional and social media characteristics.

The emergence of affective computing and sentiment analysis tools presents new avenues for exploring the subtleties of emotive influence in real-time communication networks. Future research could explore the role of mixed sentiment content and further refine models to accommodate nuanced language use and cultural variations across regions.

The paper sets a foundation for leveraging sentiment analysis in improving content dissemination strategies, both in marketing and media sectors, thereby optimizing reach in a platform as diverse and dynamic as Twitter.

This evaluation contributes to a more granular understanding of social media dynamics, aiding researchers and practitioners in crafting strategies that consider both the type of content and the emotional resonance in predicting content virality.