Searching for MH 370 using Bayes Theorem
http://www.npr.org/blogs/thetwo-way/2014/03/25/294390476/can-a-250-year-old-mathematical-theorem-find-a-missing-plane
Bayes Theorem is a mathematical formula that can be used to determine the probability of an event as it relates to the probability of a different event. The formula of Bayes Theorem is P(B|A) = P(B|A)*P(A)/P(B), relating the probability of events A and B. In addition to helping model information cascades, as we saw in class, Bayes Theorem also turns out to be an excellent tool for searching – a search algorithm based on Bayes’ Theorem can tell the probability that you will find something, given that you search in some place or area. Relative probabilities can narrow down search areas vastly and effectively.
After Malaysia Airlines flight MH 370 disappeared in March, there was an unsuccessful attempt to search for it. The plane’s potential location was extremely vague, and searching the entire Indian Ocean proved to be a daunting and almost impossible task. Metron Corp., a scientific consulting firm, handled a similar case 2009, with Air France 447. When that plane went down, both of the plane’s black boxes were damaged, and there was no way to track it. In order to pin down where it was, a Bayesian algorithm that took several factors surrounding the plane’s disappearance was employed. While the initial guess of the plane’s location was not correct, a revised version of the algorithm employed later was able to pin down the plane’s location within days. The same algorithm was employed in the search for MH 370 – however, it did not work, due to the fact that there wasn’t enough initial information for the algorithm to significantly reduce the search area. This highlights an important feature of the theorem itself – while it’s ability to determine relative probabilities can be very powerful, there needs to be accurate enough initial information for it to be of any use.