Influence Maximization for Social Good: Use of Social Networks in Low Resource Communities

TL;DR

This paper introduces a new influence maximization problem tailored for low-resource communities, proposing a POMDP-based approach and demonstrating its effectiveness through a real-world pilot with homeless youth.

Contribution

It defines the DIME problem, develops scalable POMDP algorithms, and validates their practical application in a social good context.

Findings

Algorithms successfully identified influential individuals in homeless youth networks.

Pilot study showed positive impact on social intervention efforts.

Proposed methods outperform traditional influence maximization approaches.

Abstract

This thesis proposal makes the following technical contributions: (i) we provide a definition of the Dynamic Influence Maximization Under Uncertainty (or DIME) problem, which models the problem faced by homeless shelters accurately; (ii) we propose a novel Partially Observable Markov Decision Process (POMDP) model for solving the DIME problem; (iii) we design two scalable POMDP algorithms (PSINET and HEALER) for solving the DIME problem, since conventional POMDP solvers fail to scale up to sizes of interest; and (iv) we test our algorithms effectiveness in the real world by conducting a pilot study with actual homeless youth in Los Angeles. The success of this pilot (as explained later) shows the promise of using influence maximization for social good on a larger scale.