cluster.datasets (version 1.0-1)

hardware.triads: Hartigan (1975) Triads Based on Hardware

Description

The table defines pairs of hardware objects that are most similar along with a dissimilar object. This is Table 10.1 in Chapter 10 of Hartigan (1975) on page 178.

Usage

data(hardware.triads)

Format

A data frame with 20 observations on the following 4 variables.

case

a character vector

similar.1

a factor for the first object of similar pair with levels B N P T

similar.2

a factor for the second object of similar pair with levels B F S T

odd

a factor for the different object with levels B F N P S T

Source

Hartigan, J. A. (1975). Clustering Algorithms, John Wiley, New York. SPAETH2 Cluster Analysis Datasets http://people.sc.fsu.edu/~jburkardt/datasets/spaeth2/spaeth2.html

Details

Six pieces of hardware were considered. Every possible set of three distinct pieces of hardware was examined, and a judgment was made about which two pieces were most similar. The results were reported by listing the closest pair with parentheses surrounding them, followed by the "odd" item. The hardware objects are identified as follows

1. "N" is a nail
2. "P" is a Phillips head screw
3. "B" is a bolt
4. "T" is a tack
5. "F" is a finishing nail
6. "S" is a screw

These data are used to test the triads algorithm.

References

Hartigan, J. A. (1975). Clustering Algorithms, John Wiley, New York.

Examples

data(hardware.triads)