seriation (version 1.0-1)

seriate: Seriate objects in dissimilarity matrices, matrices or arrays

Description

Tries to find an linear order for objects using data in form of a dissimilarity matrix (two-way one mode data), a data matrix (two-way two-mode data) or a data array (k-way k-mode data).

Usage

## S3 method for class 'dist':
seriate(x, method = NULL, control = NULL, \ldots)
## S3 method for class 'matrix':
seriate(x, method = NULL, control = NULL, 
    margin = c(1,2), ...)
## S3 method for class 'array':
seriate(x, method = NULL, control = NULL, 
    margin = seq(length(dim(x))), ...)

Arguments

x

the data.

method

a character string with the name of the seriation method (default: varies by data type).

control

a list of control options passed on to the seriation algorithm.

margin

a vector giving the margins to be seriated. For matrix, 1 indicates rows, 2 indicates columns, c(1,2) indicates rows and columns. For array, margin gets a vector with the dimensions to seria

...

further arguments (unused).

Value

• Returns an object of class ser_permutation.

Details

Two-way two-mode data has to be provided as a dist object (not as a symmetric matrix). Similarities have to be transformed in a suitable way into dissimilarities. Currently the following methods are implemented for dist: [object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]

Two-way two mode data are general positive matrices. Currently the following methods are implemented for matrix: [object Object],[object Object],[object Object]

For array no built-in methods are currently available.

References

P. Arabie and L.J. Hubert (1990): The bond energy algorithm revisited, IEEE Transactions on Systems, Man, and Cybernetics, 20(1), 268--274.

Z. Bar-Joseph, E. D. Demaine, D. K. Gifford, and T. Jaakkola. (2001): Fast Optimal Leaf Ordering for Hierarchical Clustering. Bioinformatics, 17(1), 22--29.

Brusco, M., Koehn, H.F., and Stahl, S. (2007): Heuristic Implementation of Dynamic Programming for Matrix Permutation Problems in Combinatorial Data Analysis. Psychometrika, conditionally accepted.

Brusco, M., and Stahl, S. (2005): Branch-and-Bound Applications in Combinatorial Data Analysis. New York: Springer.

Chen, C. H. (2002): Generalized Association Plots: Information Visualization via Iteratively Generated Correlation Matrices. Statistica Sinica, 12(1), 7--29.

Sharlee Climer, Weixiong Zhang (2006): Rearrangement Clustering: Pitfalls, Remedies, and Applications, Journal of Machine Learning Research, 7(Jun), 919--943.

Gruvaeus, G. and Wainer, H. (1972): Two Additions to Hierarchical Cluster Analysis, British Journal of Mathematical and Statistical Psychology, 25, 200--206.

Hurley, Catherine B. (2004): Clustering Visualizations of Multidimensional Data. Journal of Computational and Graphical Statistics, 13(4), 788--806.

J.K. Lenstra (1974): Clustering a Data Array and the Traveling-Salesman Problem, Operations Research, 22(2) 413--414.

W.T. McCormick, P.J. Schweitzer and T.W. White (1972): Problem decomposition and data reorganization by a clustering technique, Operations Research, 20(5), 993--1009.

See Also

criterion, solve_TSP in TSP, hclust in stats

Examples

##seriate dist
data("iris")
x <- as.matrix(iris[-5])
x <- x[sample(1:nrow(x)),]
d <- dist(x)

## default seriation
order <- seriate(d)
order

## plot
def.par <- par(no.readonly = TRUE)
layout(cbind(1,2), respect = TRUE)

pimage(d, main = "Random")
pimage(d, order, main = "Reordered")

par(def.par)

## compare quality
rbind(
        random = criterion(d),
        reordered = criterion(d, order)
     )


## seriate matrix
data("iris")
x <- as.matrix(iris[-5])

## to make the variables comparable, we scale the data
x <- scale(x, center = FALSE)

## try some methods
def.par <- par(no.readonly = TRUE)
layout(matrix(1:4, ncol = 2, byrow = TRUE), respect=TRUE)

pimage(x, main = "original data")
criterion(x)

order <- seriate(x, method = "BEA_TSP")
pimage(x, order, main = "TSP to optimize ME")
criterion(x, order)

order <- seriate(x, method="PCA")
pimage(x, order, main = "first principal component")
criterion(x, order)

## 2 TSPs
order <- c(
    seriate(dist(x), method = "TSP"),
    seriate(dist(t(x)), method = "TSP")
)
pimage(x, order, main = "2 TSPs")
criterion(x, order)

par(def.par)