ABSTRACT
In Chapter 9, we discussed the importance of location mining and the challenges we are faced with in extracting the accurate locations of social media users. In Chapter 10, we discussed our collection of TweetHood algorithms for extracting the location of Twitter users. In particular, we have described three methods that show the evolution of the algorithm currently used in TweetHood. These algorithms are as follows: (i) a simple majority algorithm with variable depth, (ii) k closest friends with variable depth, and (iii) fuzzy k closest friends with variable depth. We have also provided experimental results for the algorithms. In Chapter 11, we enhanced our location extraction algorithms and proposed a semisupervised learning method for label propagation. This algorithm was called Tweecalization. Chapter 12 described the effects of migration. People migrate from city to city, state to state, and country to country all the time. Therefore, we addressed the problem of how does one extract the location of a person when he or his friends may be continually migrating? We developed a set of algorithms that we call Tweeque. That is, Tweeque takes into account the migration effect. In particular, it identifies social cliques for location mining.
