Contextual Bandits for Maximizing Stimulated Word-of-Mouth Rewards

2026-06-13

AI Summary

This study addresses the challenge of accurately identifying and incentivizing users in social networks who are most likely to generate substantial spillover effects, thereby maximizing the returns from incentivized word-of-mouth campaigns. The authors propose a contextual multi-armed bandit framework that integrates individual-level heterogeneity in spillover effects, enabling online learning of personalized spillover probabilities to dynamically rank and select target users. By introducing fine-grained spillover modeling into the multi-armed bandit paradigm for the first time, the method significantly improves the accuracy of identifying top-k high-value users on real-world social network data and achieves higher cumulative rewards compared to baseline approaches that ignore individual differences.

Abstract

Stimulated word-of-mouth is a strategy that promotes information sharing through prompts or incentives. Optimizing stimulated word-of-mouth through social networks requires identifying and targeting connected users who are most susceptible to spillover, a phenomenon where the influence of recommendations extends beyond the immediate audience to impact their connected users. The probability of spillover varies across individuals, and their connections, leading to heterogeneity. Understanding and accurately estimating the spillover probabilities among users in social networks is crucial for improving the effectiveness of stimulated word-of-mouth. To address this, we present a novel contextual multi-armed bandit framework that learns individual spillover probabilities and ranks connected users to maximize rewards from stimulated word-of-mouth. Experiments on real-world network datasets demonstrate that accounting for spillover heterogeneity enhances the targeting precision of top-$k$ connected users, boosting rewards and outperforming baseline methods that do not learn individual spillover effects.

Problem

• stimulated word-of-mouth
• spillover effect
• social networks
• heterogeneity
• targeting

Innovation

• contextual bandits
• spillover effects
• stimulated word-of-mouth
• heterogeneous influence
• social network targeting