0
$\begingroup$

While there is lot of theory and probability in the background to understand, I wanted to know if there are any resources/quick pointers on what to consider while modeling a problem using Bayesian Networks.

Specifically, I had the following questions: - what kind of priors to use? - what do we learn from the data? and how to do prediction?

Is there a blog/document that explains these from the ground-up with simple use-cases? (eg: modeling user-click behavior, weather forecasting, diagnosis etc). I am looking for examples which are a little detailed beyond toy-examples (a bit close to real life situations) used in courses.

Any help is appreciated.

Improve this question
$\endgroup$
0

3 Answers 3

Reset to default
2
$\begingroup$

This paper explains the design stages of a Bayesian Network based on a real-life data-set and for a real-life application. It touches on pre-processing data, how to learn the probability space (MLE - which is simple), how to learn the DAG's (which I refer to as causal discovery), what to take into account while learning the DAGs, etc.

I am happy to provide more information if you have questions.

Improve this answer
$\endgroup$
1
$\begingroup$

Your best bet would be reading research papers that use those models for some task. I assume that this task will most often be classification.

I know of one book about applications of BNs:

  • Bayesian network technologies: applications and graphical models. Mittal, Ankush and Kassim, Ashraf, 2007, IGI

Concerning Markov networks, you can look for applications using Markov logic. ML is some modeling language for Markov networks, and there exist some applications, for example:

  • Goal-based person tracking using a first-order probabilistic model T Geier, S Reuter, K Dietmayer, S Biundo - Proceedings of the Ninth UAI … Workshop, 2012
  • Event modeling and recognition using markov logic networks. SD Tran, LS Davis - Computer Vision–ECCV 2008, 2008 - Springer
  • Goal Recognition with Markov Logic Networks for Player-Adaptive Games. EY Ha, JP Rowe, BW Mott, JC Lester - AIIDE, 2011 - aaai.org
  • Recognizing Multi-Agent Activities from GPS Data. A Sadilek, HA Kautz - AAAI, 2010 - aaai.org
Improve this answer
$\endgroup$
0
$\begingroup$

You can check out the Coursera course on Probabilistic Graphical Models. Though there isn't any future session announced, you can still watch the videos

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.