A few years ago I finally finished my bachelor's degree in statistics that I had been working on for a number of years. (It took me a while as I've been working full time and also because the University of Helsinki is in Helsinki while I've been in London since 2011.) My thesis was on PLSA, which is a latent topic model for clustering text documents.
Download the thesis here as PDF. I decided to publish it online in case it proves interesting to people interested in statistical models. The thesis was originally written in Finnish but I've translated it into English for publication.
Hope y'all find it interesting.
