[R] dissimilarity matrix using SAX distance

[AURORA GONZALEZ VIDAL]

Dear all,

I'm trying to cluster some data using SAX distance that was described in
the paper "a symbolic representation of time series with implications for
streaming algorithms" http://www.cs.ucr.edu/~eamonn/SAX.pdf

Once I have my data in matrix format, which function can I use to compute
the dissimilarity matrix? There are several ones to compute the distance
between two SAX data series

diss.MINDIST.SAX(x, y, w, alpha, plot=TRUE)
Func.dist(x, y, matrix, n)

but it is very slow when I try to fill the matrix with two loops and I
really think there should be already any implentation. Do you have any
idea?

I already convert the data into a series of "a",  "b", "c", ... etc data
so I would appreciate either the directo computation of the sax matrix
using my raw data OR using the data already converted to SAX format.

Thank you for any suggestion!


------
Aurora González Vidal
Phd student in Data Analytics for Energy Efficiency

Faculty of Computer Sciences
University of Murcia

@. aurora.gonzalez2 en um.es
T. 868 88 7866
www.um.es/ae

	[[alternative HTML version deleted]]