Ranked Nodes Method Concerning MATLAB Functions

MATLAB functions to support the elicitation of parameters of the ranked nodes method.

The zip folder contains MATLAB files related to the paper “Advancing Construction of Conditional Probability Tables of Bayesian networks with Ranked Nodes Method” by Pekka Laitila and Kai Virtanen, Department of Mathematics and Systems Analysis, Aalto University, Finland. The paper presents a framework to support the determination of parameters of the ranked nodes method that is used to construct conditional probability tables for Bayesian networks by expert elicitation.

There are15 MATLAB function files in the zip folder. Three main files are:

1. feasibilityCheck_singleWE.m
2. feasibilityCheck_partitionedWE.m
3. plotProbDstrs_DR_scens.m

The function feasibilityCheck_singleWE.m implements the computational procedure of Step 3 of the elicitation framework presented in the paper.

The function feasibilityCheck_partitionedWE.m implements the computational procedure of the modified Step 3 of the elicitation framework concerning the use of a partitioned weight expression.

The function plotProbDstrs_DR_scens.m plots probability distributions concerning both the “normal” Step 4 of the elicitation framework as well as the modified Step 4 concerning the use of a partitioned weight expression.

The other 12 files are supporting functions for the three main files. Further information on the files is provided in the documents Notes1.pdf and Notes2.pdf included in the zip folder.