Optimizing Opinions with Stubborn Agents
提出一种基于观点动力学模型的算法，用于在社交网络中放置说服代理人以最大化影响力，可调节观点均值或方差来改变极化程度，在真实Twitter网络模拟中效果显著。
How can we place persuasion agents in a social network to influence a population? In “Optimizing Opinions with Stubborn Agents,” David Scott Hunter and Tauhid Zaman present an algorithm based on opinion dynamics models that shows where to place these agents in a network for maximum persuasive effect. Using this algorithm, one can shape opinions in a variety of interesting ways. For instance, one can use the agents to maximize the opinion mean in order to increase support for an issue. More interestingly, one can use the algorithm to shape the opinion variance in order to decrease, or even increase, the polarization in a network. Simulations on a variety of real Twitter networks showed that, with their algorithm, a small number of strategically placed agents can create significant opinion shifts.