Topic diffusion in blogs
		n Topic =  keyword n Need to track relevant words w.r.t. time ¡ tf  cidf (cumulative idf); corpus is a moving window n Find three distributions of topics ¡ Chatter: topics continuously discussed (e.g., alzheimers) ¡ Spike: topic exhibiting a usage spike, then inactivity (e.g., chibi) ¡ Spiky Chatter: Topics (e.g., microsoft) n Overlay of above two types (multiple spikes possible) n Spike removal possible with spike model