Network-based intervention strategies to reduce violence among homeless
Network-based intervention strategies to reduce violence among homeless
复制标题
DOI：
10.1007/s13278-019-0584-8
复制
发表时间：
2019-07-27
影响因子：
2.8
通讯作者：
Prasanna, Viktor K.
中科院分区：
文献类型：
--
作者：
Srivastava, Ajitesh;Petering, Robin;Prasanna, Viktor K.
暴力是一种严重影响无家可归青年的现象，由于许多促成因素，如创伤性童年经历、参与犯罪活动以及因街头占有权而接触犯罪者，无家可归青年遭受暴力的风险更大。减少这一人群中的暴力行为是确保这些人能够安全和成功地摆脱无家可归状态并过上长期有所作为的生活所必需的。由于暴力行为的复杂性，减少这一人群暴力行为的干预措施难以实施。然而，基于同伴的干预方法可能是一种有价值的方法，因为以前的研究表明，与暴力行为更严重的人互动的人更有可能暴力，这表明暴力具有传染性。我们提出了不确定的选民模型来表示复杂的过程中的暴力扩散的社会网络，捕捉不确定性的链接和时间的暴力扩散发生。假设这个模型，我们定义的暴力最小化问题的任务是选择一个预先定义的数量的个人进行干预，使网络中的暴力个人的预期数量在给定的时间范围内最小化。我们还将问题扩展到概率设置，其中将个人转换为非暴力的成功概率是对他们进行干预的单位数量的函数。我们提供了算法，为这两种情况下找到最佳的干预策略。我们证明，我们的算法在减少暴力方面的表现明显优于基于流行中心性措施的干预措施。最后，我们使用我们的概率干预的最佳算法招募同龄人在无家可归的青年收容所作为试点研究。我们在干预前后进行的调查显示，暴力行为大幅减少。
Violence is a phenomenon that severely impacts homeless youth who are at an increased risk of experiencing it as a result of many contributing factors such as traumatic childhood experiences, involvement in delinquent activities, and exposure to perpetrators due to street tenure. Reducing violence in this population is necessary to ensure that the individuals can safely and successfully exit homelessness and lead a long productive life. Interventions to reduce violence in this population are difficult to implement due to the complex nature of violence. However, a peer-based intervention approach would likely be a worthy approach as previous research has shown that individuals who interact with more violent individuals are more likely to be violent, suggesting a contagious nature of violence. We propose uncertain voter model to represent the complex process of diffusion of violence over a social network that captures uncertainties in links and time over which the diffusion of violence takes place. Assuming this model, we define violence minimization problem where the task is to select a predefined number of individuals for intervention so that the expected number of violent individuals in the network is minimized over a given time frame. We also extend the problem to a probabilistic setting, where the success probability of converting an individual into nonviolent is a function of the number of units of intervention performed on them. We provide algorithms for finding the optimal intervention strategies for both scenarios. We demonstrate that our algorithms perform significantly better than interventions based on popular centrality measures in terms of reducing violence. Finally, we use our optimal algorithm for probabilistic intervention to recruit peers in a homeless youth shelter as a pilot study. Our surveys before and after the intervention show a significant reduction in violence.