Recursive fact-finding
Abstract
As we discussed earlier in Chapter 5, expectation maximization (EM) is an iterative algorithm that provides the maximum likelihood estimation when the iteration converges. However, running the iterative algorithm is not necessarily efficient for the streaming data, especially when the estimation parameter θ remains stable over time. This observation motivates the development of a real-time version of the EM algorithm for streaming data to achieve a better tradeoff between the estimation accuracy and running time. In this chapter, we present a recursive fact-finding model developed based on EM approach that allows the applications to update their estimation on the fly as data stream in.
Keywords ...
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.
