UnBBayes is a probabilistic network framework written in Java. It has both a GUI and an API with inference, sampling, learning and evaluation. It supports Bayesian networks, influence diagrams, MSBN, OOBN, HBN, MEBN/PR-OWL, PRM, structure, parameter and incremental learning.
- Bayes Net (or Bayesian network)
- Influence Diagrams (or Decision Graphs)
- MSBN (Multiple-Sectioned Bayesian Network)
- OOBN (Object-Oriented Bayesian Network)
- MEBN (Multi-Entity Bayesian Network)
- PR-OWL (Probabilistic Web Ontology Language)
- PR-OWL 2
- UMP-ST (Uncertainty Modeling Process for Semantic Technologies)
- Sampling and Simulation (e.g. Monte Carlo, Gibbs)
- Approximate Inference
- Data Mining
- Plugin Support
- PRM (Probabilistic Relational Model)
The best open source package for working with PGMs. Supports decision nodes and utility nodes (unlike SamIam). Is open source (unlike Geany and SamIam). Many plugins for different things. However, I do wish that it implemented different inference algorithms, such as recursive conditioning and loopy belief propapagation, and had the ability to perform sensitivity analysis.
UnBBayes is a free and easy to use tool for Bayesian Networks. I am using it on my Master Thesis.
Just tried the newer (plugin based) version of unbBayes and glad to see the refinements. Ability to see the complete MTheory and do adjustments by dragging & resizing MFRag panels is cool.
I've been using UnBBayes for several years now, and it has come a long way. I am happy to have an implementation of MEBN. The changes over the past couple of years -- the GUI improvements, the plugin framework, the bug fixes have made it much better. I am really looking forward to seeing PR-OWL 2.0.
UnBBayes is a great framework for modeling with both traditional bayesian networks as well as provides an excellent way to model MEBN's. There has been a lot of improvements with the new version , especially with the plugin framework and GUI and with the other changes over the past months ,it is turning out to be a really great tool.