Description:

The AMIDST toolbox offers a collection of scalable and parallel algorithms for inference and learning of hybrid static and dynamic Bayesian networks from streaming data.

For instance, AMIDST provides parallel multi-core implementations of Bayesian parameter learning algorithms, using variational message passing based algorithms. Inference can be effectively carried out with parallel Monte Carlo techniques. Additionally, AMIDST efficiently leverages existing functionalities and algorithms by interfacing to software tools such as Weka, Moa, HUGIN and R.

Example of well-known supported models:

• Static models: Naive Bayes, TAN, AODE, Gaussian Discriminant Analysis, Latent Classification Models, Gaussian Mixtures, Bayesian Linear Regression, Factor analysis, Mixture of Factor Analysis, ...

• Dynamic models: Dynamic Naive Bayes, Dynamic Latent Classification Models, Hidden Markov Model (HMM), Kalman Filter (KF), Switching KF, Factorial HMM, Auto-regressive HMM, Input-Output HMM, ...

Changes to previous version:

Initial Announcement on mloss.org.

BibTeX Entry: Download

Corresponding Paper BibTeX Entry: Download

Supported Operating Systems: Platform Independent

Data Formats: Arff

Tags: Approximate Inference, Bayesian Networks, Data Streams, Multi Core, Bayesian Learning, Hidden Markov Models, Importance Sampling, Maximum Likelihood, Parallelisation, Varational Message Passing, Kalma

Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.