next up previous
Next: 2.1.1 Data Format Up: 2 Software Previous: 2 Software

2.1 Design

All of the tools read their input (if any) from standard input (stdin), and write their output (if any) to standard output (stdout).

The programs are written in ANSI C and C++, using the LEDA library. They could easily be extended to be part of some larger data mining application or set of shell tools.

Points are specified as two real valued numbers, that is, a point in the real plane. Clusters are specified by the non-negative integers, the cluster -1 denotes the set of outlier points. Lastly, the points in a dataset may have ``annotation'', which is simply arbitrary data associated with the point.


Kevin Pulo