Measuring user influence on Twitter: A survey (1508.07951v2)

Abstract: Centrality is one of the most studied concepts in social network analysis. There is a huge literature regarding centrality measures, as ways to identify the most relevant users in a social network. The challenge is to find measures that can be computed efficiently, and that can be able to classify the users according to relevance criteria as close as possible to reality. We address this problem in the context of the Twitter network, an online social networking service with millions of users and an impressive flow of messages that are published and spread daily by interactions between users. Twitter has different types of users, but the greatest utility lies in finding the most influential ones. The purpose of this article is to collect and classify the different Twitter influence measures that exist so far in literature. These measures are very diverse. Some are based on simple metrics provided by the Twitter API, while others are based on complex mathematical models. Several measures are based on the PageRank algorithm, traditionally used to rank the websites on the Internet. Some others consider the timeline of publication, others the content of the messages, some are focused on specific topics, and others try to make predictions. We consider all these aspects, and some additional ones. Furthermore, we include measures of activity and popularity, the traditional mechanisms to correlate measures, and some important aspects of computational complexity for this particular context.

Measuring User Influence on Twitter: An Analytical Survey

The paper entitled "Measuring user influence on Twitter: A survey" by Fabián Riquelme and Pablo González-Cantergiani systematically explores various user influence metrics within Twitter's social network. The emphasis is on identifying the most suitable measures to ascertain the influence, activeness, and popularity of users on Twitter, thus providing a comprehensive analysis tailored to cater to the complex interplay of the platform.

Methodological Overview

This survey categorizes influence measures into three distinct groups: Activity Measures, Popularity Measures, and Influence Measures. Each subset caters to different aspects of a user's interaction on Twitter, reflecting specific attributes like frequency of posts, number of followers, extent of content dissemination, etc.

1. Activity Measures: These include basic metrics such as tweet count and involve computation of a user's visible actions including tweets, retweets, mentions, and likes over a predetermined time frame. They are particularly beneficial for assessing a user’s engagement level without evaluation of the audience's response.
2. Popularity Measures: Usually contingent on the extent of a user’s network, popularity is quantified by follower counts and followee interactions. Several methods leverage both direct counts and derived metrics like the 'FollowerRank' or variations of the 'in-degree' centrality measure to calculate perceived popularity.
3. Influence Measures: These measures aim to quantify how a user's actions affect other users within the network. The implementation of PageRank adaptations, which traditionally evaluate web page authority, has been extended to Twitter to assess how tweets propagate through network nodes.

Implications and Observations

The survey brings forth the diversity and complexity inefficiencies inherent in influence measurement approaches on online platforms like Twitter. The choice of metric heavily influences the classification of users and is typically contingent on the availability of detailed interaction data.

• Techniques like PageRank are notable for their recursive nature, assigning users a measure of importance based on the influence of those they influence.
• The computational complexity of these measures varies, often demanding significant resources depending on the metric. For instance, understanding 'Influence Rank' or 'DIS' can necessitate intensive data processing techniques that could be infeasible for real-time analysis.
• Measures like Retweet Impact and Mention Impact provide simpler alternatives, emphasizing specific Twitter interactions without relying on comprehensive API data, thereby facilitating immediate analysis.

Future Directions

With the constraints imposed by Twitter APIs, the feasibility of detailed follower-followee data scrubbing is limited, necessitating innovative approaches to influence measurement that can offer robust, dynamic feedback. As the field evolves:

• There is a call for integration of multi-dimensional analysis frameworks that surpass basic engagement metrics, incorporating dimensions such as sentiment analysis and topical relevance to align influence with content impact.
• Furthermore, the potential for real-time influence dynamics tracking puts forward challenges for developing graph algorithms that can process data streams efficiently.

Conclusion

This paper serves as a baseline reflection on the measurement of user influence within Twitter, compiling an array of methodologies while acknowledging the challenges and limitations of capturing the essence of 'influence' on evolving social networks. By offering a structured survey, it provides valuable insights and acts as a springboard for future research to explore more comprehensive and computationally manageable ways to understand and leverage user influence within digital ecosystems.