ICA / Independent Component Analysis, is one of the decomposition methods that is capable of decomposing multivariate signals into components, based on the assumption that the source components are statistically independent from each other, and follow a non-gaussian distribution.
Sorry for the shilling, but here’s my upcoming project:
Please register your email address if you’re interested in it.
Edit: Sadly by the request of Coursera itself, I’ve removed this particular github repository. Coursera is doing the right thing though, don’t blame them.
I applied and followed Machine Learning course which is taught by Andrew Ng and offered freely online on Coursera (https://www.coursera.org/course/ml). However since the Octave installed on my computer doesn’t work properly (e.g. plotting functionality doesn’t work at all, etc), I decided to use python instead.