Skip to main content

Table 1 Parameters defining the simulation scenarios as used in generating data and labels

From: Consensus clustering for Bayesian mixture models

Scenario

N

\(P_s\)

\(P_n\)

K

\(\Delta \mu\)

\(\sigma ^2\)

\(\pi\)

2D

100

2

0

5

3.0

1

\((\frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5})\)

Small N, large P

50

500

0

5

1.0

1

\((\frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5})\)

Irrelevant features

200

20

100

5

1.0

1

\((\frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5}, \frac{1}{5})\)

  1. \(\Delta \mu\) is the distance between neighbouring cluster means within a single feature. The number of relevant features (\(P_s\)) is \(\sum _p \phi _p\), and \(P_n = P - P_s\)