Conditional Random Fields and Graphical Models

I am currently reading into the topic of Conditional Random Fields, Hidden Markov Models in particular and Graphical Models in general. A perfect tutorial on this topics is provided by Sutton and McCallum [1]. It is more than readworthy and gives

• a basic introduction to relational learning for graphical models
• an overview on CRF and HMM as well as their application
  
• Generative vs. descriptive models (i.e. Naive Bayes vs. Linear Regression)
• Parameter estimation, backward-forward estimation and application of gneral and linear chain CRF's
• Application of general CRF's (i.e. skip chain CRF) for information extraction

A must read for everybody interested in information extraction and relational learning state-of-the-art

[1] Introduction to Conditional Random Fields for Relational Learning

Charles Sutton and Andrew Mccallum MIT Press, (2006)

What is it for?

This blog gathers stuff I read on knowledge relationship discovery or in more detail on machine learning, natural language processing, semantic technologies and all those stuff.
Entries will be mostly used for personal recall of stuff i stumbled over in the past, so all entries may not be self contained and may require some know-how in the above fields. However, if you can take away something you are welcome