Interacting with Users in Social Networks: The Follow-back Problem
Abstract
Imagine in a social network that an agent wants to form a connection with a target user(s) by interacting with the friends of the target(s). Because forming a connection is known as following in social networks such as Twitter, we refer to this as the follow back problem. The friends of the target user(s) form a directed graph which we refer to as the friend's graph. The agents goal is to get the target to follow him, and he is allowed to interact with the target and the targets friends. To understand what features impact the probability of an interaction resulting in a follow-back, we conduct an empirical analysis of several thousand interactions in Twitter. We build a model of the follow-back probabilities based upon this analysis which incorporates features such as the friend and follower count of the target and the neighborhood overlap of the target with the agent. We then use this model to solve the follow-back problem. We find optimal policies for simple network topologies such as directed acyclic graphs and single cycle graphs. For arbitrary directed graphs we develop heuristics based upon a graph score measure we define as the followback score. We show through simulation that these heuristic policies perform well on real Twitter graphs.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 02, 2016
- Accession Number
- AD1033681
Entities
People
- Krishnan Rajagopalan
Organizations
- Massachusetts Institute of Technology