Taboo Learning

Context

Taboo Learning refers to structural learning that implements the so-called Taboo search for Bayesian networks. This method is particularly useful for refining networks built by human experts or for updating a network learned from a previous dataset.

Usage

Beyond taking into account the a priori knowledge represented by a given network and the specified equivalent number of cases, the starting point of Taboo is the current network, not the fully unconnected network (no arcs), as is the case for SopLEQ and Taboo Order. Furthermore, Fixed Arcs remain unchanged and Forbidden Arcs are taken into account. Temporal Indices are also taken into account. You can define the size of the Taboo list as well as the maximum number of parents and children. Unless these options are checked, no constraints are used. Furthermore, you can specify whether to keep the existing network structure or not.