DOCTOR-DROID.BLOGSPOT.COM
Dr. DroidSunday, April 25, 2010. Analyzied result of mine. After I draw some graphs, I found that my result is not consistent with Andrew's. So I changed my algorithm so that HMM can accept different size of sequence. Currently, I normalize the size and then calculates log-likelihood). After that, I could get reasonable result. You can download graph here. Presentation outline (maybe for report/paper too? 1: Hook (X stroke rate, Y death rate. Obviously rehab is important). 2a: show video of select exercises.
http://doctor-droid.blogspot.com/


