From bfaswiki
Jump to: navigation, search

This is a list of toolboxes based on the theory of belief functions that facilitates their handling. If you want to share your software contact us: [1].

Links Languages Comments
Fast Mobius Transform

for PC for Mac

for Unix
Matlab A set of functions written in Matlab developed by Philippe Smets. FMT = Fast Mobius Transforms.
TBMLAB Matlab A full demonstrator for the TBM. In folder Documents, read the READ_ME file, developed by Philippe Smets
DST Matlab Easy to test (see test.m). All elements of the power sets are coded using Smets codes of Mobius transform. Developed by Arnaud Martin.
Many links Matlab Many softwares for belief clustering and belief classifiers developed by Thierry Denoeux.
General Framework Matlab Easy to test (see test.m). Only focal elements are coded, works for power set and hyper power set developed by Arnaud Martin.
Referee functons Java Referee functions toolbox developed by Frédéric Dambreville.
IPP Matlab
Matlab and R The IPP Toolbox is a collection of methods for uncertainty quantification and propagation using Dempster-Shafer Theory and imprecise probabilities.
iBelief R

The R package ibelief aims to provide some basic functions to implement the theory of belief functions, and it has included many features such as:

  1. Fast Mobius Transformation to convert any of the belief measures (such as basic belief assignment, credibility, plausibility and so on) to another type;
  2. Some commonly used combination rules including DS rule, Smets’ rule, Yager’s rule, DP rule, PCR6 and so on;
  3. Some rules for making decisions;
  4. The discounting rules in the theory of belief functions;
  5. Different ways to generate random masses.

The stable version of package ibelief could be found on <a title="ibelief" href="" target="_blank">CRAN</a>. The following command can be used in R to install the package:
install.packages(‘ibelief’, dependencies = TRUE)

Belief R

The belief R package is a collection of basic tools to handle belief functions. It is currently limited to finite spaces and encode mass assignments both in extensive form (all elements of the power set) and restricted form (only focal elements).

Belief Function Machine (BFM)
Matlab The Belief Function Machine (BFM), that was developed at Univ. of Kansas, with the advice/supervision of Philippe Smets, Thierry Denoeux, and Prakash P. Shenoy, by Phan Hong Giang (a former PhD student, now on the faculty of George Mason Univ.), and funded by Raytheon Missile Systems, Tucson, AZ. BFM can find marginals, translate marginals to probabilities, do sensitivity analysis, deal with conditionals, manipulate the join tree, etc.