Anonymized Local Privacy

In this paper, we introduce the family of Anonymized Local Privacy mechanisms. These mechanisms have an output space of three values "Yes", "No", or "$\perp$" (not participating) and leverage the law of large numbers to generate linear noise in the number of data owners to protect privacy both before and after aggregation yet preserve accuracy. We describe the suitability in a distributed on-demand network and evaluate over a real dataset as we scale the population.