Ego-centered network intervention via weak-tie exclusion

Social distancing has been widely adopted to curb infectious disease outbreaks, but large-scale lockdowns and individual quarantines impose heavy social, psychological, and economic costs. This motivates the design of more sustainable, ego-centered contact-reduction strategies that operate at the individual level. In networks, distancing can be modeled by edge removal, but existing approaches rely on global topological information inaccessible to individuals. We assume behavioral changes are guided by local observations: each person selectively avoids contacts based on simple, locally accessible information. Inspired by weak-tie theory and the pronounced community structure of social networks, we propose a two-stage hybrid intervention whereby individuals first target low-degree neighbors for removal and subsequently eliminate ties with the fewest common neighbors. This ego-centered approach fragments community structures and isolates peripheral components, enabling effective epidemic suppression. Compared with strategies targeting highly connected nodes, our method achieves better outcomes while reducing social costs. These results highlight that interventions grounded in individual-level decisions and local tie information can inform practical social distancing policies without requiring global network knowledge.