mmb0.13.3 package

Arbitrary Dependency Mixed Multivariate Bayesian Models

bayesCaret

Provides a caret-compatible wrapper around functionality for classific...

bayesComputeMarginalFactor

Compute a marginal factor (continuous or discrete random variable).

bayesComputeProductFactor

Computes one single factor that is needed for full Bayesian inferencin...

bayesConvertData

Convert data for usage within Bayesian models.

bayesFeaturesToSample

Transform a collection of Bayesian features back to a sample.

bayesInferSimple

Perform simple (network) Bayesian inferencing and regression.

bayesProbability

Full Bayesian inferencing for determining the probability or relative ...

bayesProbabilityAssign

Assign probabilities to one or more samples, given some training data.

bayesProbabilityNaive

Naive Bayesian inferencing for determining the probability or relative...

bayesProbabilitySimple

Assign a probability using a simple (network) Bayesian classifier.

bayesRegress

Perform full-dependency Bayesian regression for a sample.

bayesRegressAssign

Regression for one or more samples, given some training data.

bayesRegressSimple

Perform simple (network) Bayesian regression.

bayesToLatex

Create a string that can be used in Latex in an e.g. equation-environm...

centralities

Given a neighborhood of data, computes the similarity of each sample i...

checkBayesFeature

Validate a Bayesian feature using some sanity checks.

conditionalDataMin

Segment data according to one or more random variables.

createFeatureForBayes

Create a Bayesian feature by name and value.

discretizeVariableToRanges

Discretize a continuous random variable to ranges/buckets.

distance

Given a neighborhood of data and two samples from that neighborhood, c...

estimatePdf

Safe PDF estimation that works also for sparse random variables.

getDefaultRegressor

Get the system-wide default regressor.

getMessages

Get a boolean indicating whether messages are enabled system-wide.

getProbForDiscrete

Get a probability of a discrete value.

getRangeForDiscretizedValue

Get the range-/bucket-ID of a given value.

getValueKeyOfBayesFeatures

Obtain the type of the value of a Bayesian feature.

getValueOfBayesFeatures

Obtain the value of a Bayesian feature.

getWarnings

Get a boolean indicating whether warnings are enabled system-wide.

make.varClosure

Creates a closure over a variable and returns its getter and setter.

neighborhood

Given Bayesian features, returns those samples from a dataset that exh...

sampleToBayesFeatures

Transform an entire sample into a collection of Bayesian features.

setDefaultRegressor

Set a system-wide default regressor.

setMessages

Enable or disable messages system-wide.

setWarnings

Enable or disable warnings system-wide.

vicinities

Segment a dataset by each row once, then compute vicinities of samples...

vicinitiesForSample

Segment a dataset by a single sample and compute vicinities for it and...

Supports Bayesian models with full and partial (hence arbitrary) dependencies between random variables. Discrete and continuous variables are supported, and conditional joint probabilities and probability densities are estimated using Kernel Density Estimation (KDE). The full general form, which implements an extension to Bayes' theorem, as well as the simple form, which is just a Bayesian network, both support regression through segmentation and KDE and estimation of probability or relative likelihood of discrete or continuous target random variables. This package also provides true statistical distance measures based on Bayesian models. Furthermore, these measures can be facilitated on neighborhood searches, and to estimate the similarity and distance between data points. Related work is by Bayes (1763) <doi:10.1098/rstl.1763.0053> and by Scutari (2010) <doi:10.18637/jss.v035.i03>.

  • Maintainer: Sebastian Hönel
  • License: GPL-3
  • Last published: 2020-09-23