Understanding physical dependencies
Social meets cyber-physical
Abstract
All of the models assumed up to point make two important assumptions: it is assumed (1) the claims are all independent and (2) the sources have the opportunity to observe all claims. However, the above assumptions may not always hold in various kinds of social sensing applications where physical dependencies exist among sources and claims. In this chapter, we present a new maximum likelihood estimation (MLE) model that is aware of such physical dependencies and derived the corresponding expectation maximization (EM) approaches to remove the above two assumptions from the basic MLE model. It turns out the new model and EM scheme outperforms the regular EM approach ...
Get Social Sensing now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
