# Factorization Bandits for Online Influence Maximization

## TL;DR

This paper introduces a novel online influence maximization method leveraging network assortativity and factorization of influence probabilities, significantly improving learning efficiency and effectiveness in social networks.

## Contribution

It proposes a factorization-based online learning algorithm utilizing network assortativity, which is a new approach in influence maximization.

## Key findings

- Significant regret reduction demonstrated in experiments.
- Effective estimation of influence probabilities through latent factorization.
- Empirical results on real-world networks validate the approach.

## Abstract

We study the problem of online influence maximization in social networks. In this problem, a learner aims to identify the set of "best influencers" in a network by interacting with it, i.e., repeatedly selecting seed nodes and observing activation feedback in the network. We capitalize on an important property of the influence maximization problem named network assortativity, which is ignored by most existing works in online influence maximization. To realize network assortativity, we factorize the activation probability on the edges into latent factors on the corresponding nodes, including influence factor on the giving nodes and susceptibility factor on the receiving nodes. We propose an upper confidence bound based online learning solution to estimate the latent factors, and therefore the activation probabilities. Considerable regret reduction is achieved by our factorization based online influence maximization algorithm. And extensive empirical evaluations on two real-world networks showed the effectiveness of our proposed solution.