Accelerating Social Contagion: Transforming SIS Models with External Recruitment

TL;DR

Information in social networks doesn't just spread through word-of-mouth; it is often sparked and sustained by external forces like advertising. This paper reformulates the classic SIS (Susceptible-Infected-Susceptible) epidemic model to include a constant "recruitment" rate. By solving the underlying differential equations in closed form, the authors provide a powerful toolkit for optimizing marketing and political campaigns under real-world budget constraints.

Problem & Motivation: Beyond "Organic" Spreading

Most baseline models for information diffusion assume an organic process: you learn about a product because your friend told you (Susceptible-Infected contact). However, this ignores the reality of modern campaigning where media intervention (ads, billboards, sponsored content) directly "infects" the population regardless of their social circle.

The authors identify a gap in literature:

1. Passive Theory: Standard models are too slow as they rely solely on contact.
2. Complex Computation: Earlier studies using time-varying controls (Pontryagin's Principle) are mathematically "black boxes" that don't allow for simple, intuitive calculations of system evolution.

The core insight here is that by injecting a constant external control $u$, we don't just add more "infected" nodes; we create a catalytic effect where those new nodes then go on to infect others organically.

Methodology: The Mathematics of Recruitment

The authors modify the standard SIS rate equation: $\dot{i}(t)=\beta i(t)(1-i(t))-\gamma i(t)+u(1-i(t))$

Where:

• $\beta$: Message diffusion rate.
• $\gamma$: Recovery/forgetting rate.
• $u$: External recruitment rate (The Control).

By applying partial fraction expansion, they derive a Closed-Form Solution for $i(t)$, which is a rare and valuable find in non-linear dynamics. This allows a campaign manager to predict exactly what percentage of the population will be "infected" (aware) at any deadline $T$.

The figure above illustrates how different levels of control (u) change the system's equilibrium. Note that as u increases, the limit of awareness in the population moves significantly higher.

Experiments & Results: Is it Worth the Cost?

The paper evaluates two primary scenarios:

1. Weighted Cost: Balancing the reward of awareness against the expense of the campaign. The authors find the cost function to be convex, meaning there is a mathematically "perfect" amount of advertising spend that maximizes ROI.
2. Fixed Budget: Maximizing awareness given a hard limit on resources.

This chart demonstrates the relationship between the budget (B) and the final awareness (i(T)). It serves as a lookup table for campaign planners to determine the funding required for a target penetration rate.

Key Findings:

• Equilibrium Shift: External recruitment doesn't just speed up the process; it permanently raises the "steady state" of awareness in the population (Endemic State).
• Monotonicity: Awareness is a monotonically increasing function of control—meaning in a fixed budget scenario, it is always optimal to use the full budget early and consistently.

Critical Analysis & Conclusion: Future Outlook

This work provides a rigorous foundation for quantitative marketing. By moving from "rules of thumb" to "closed-form equations," it allows for predictable campaign outcomes.

Takeaway: The "Direct Recruitment" term is a simple but effective proxy for advertising. It proves that external intervention acts as a force multiplier for organic word-of-mouth.

Limitations & Future Work:

• Network Homogeneity: The model assumes an average degree $k$ for all nodes. In reality, "Influencers" (high-degree nodes) would change the dynamics of $\beta$.
• Constant Rate: Future research could explore "burst" advertising (high $u$ at the start) versus sustained campaigns to see which captures the market faster.

Ultimately, this paper bridges the gap between epidemic theory and economic strategy, providing a clear roadmap for anyone looking to "infect" a social network with a new idea.