Actively Learning to Attract Followers on Twitter

TL;DR

This paper explores an online active learning approach using a contextual bandit framework to understand how to attract followers on Twitter through retweeting, revealing insights into dynamic social behaviors.

Contribution

It introduces a novel online learning method for follower acquisition on Twitter, contrasting passive data analysis with active, real-time learning.

Findings

Aggregating experience across agents can reduce prediction accuracy.

Twitter community responses are non-stationary over time.

Active online learning yields deeper insights into follower attraction.

Abstract

Twitter, a popular social network, presents great opportunities for on-line machine learning research. However, previous research has focused almost entirely on learning from passively collected data. We study the problem of learning to acquire followers through normative user behavior, as opposed to the mass following policies applied by many bots. We formalize the problem as a contextual bandit problem, in which we consider retweeting content to be the action chosen and each tweet (content) is accompanied by context. We design reward signals based on the change in followers. The result of our month long experiment with 60 agents suggests that (1) aggregating experience across agents can adversely impact prediction accuracy and (2) the Twitter community's response to different actions is non-stationary. Our findings suggest that actively learning on-line can provide deeper insights about how to attract followers than machine learning over passively collected data alone.