https://github.com/deric/handl-data-generators

Clustering data generators
https://github.com/deric/handl-data-generators

Last synced: about 1 year ago
JSON representation

Clustering data generators

• Host: GitHub
• URL: https://github.com/deric/handl-data-generators
• Owner: deric
• License: lgpl-3.0
• Created: 2015-05-26T15:39:08.000Z (about 11 years ago)
• Default Branch: master
• Last Pushed: 2015-11-23T12:33:35.000Z (over 10 years ago)
• Last Synced: 2025-02-09T20:42:19.842Z (over 1 year ago)
• Language: C
• Size: 1.13 MB
• Stars: 3
• Watchers: 3
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.md

Awesome Lists containing this project

README

          
# J. Handl: data generators

Syntetic data generators for creating datasets with Gaussian distribution. The code was taken from [official website](http://personalpages.manchester.ac.uk/mbs/Julia.Handl/generators.html) and slightly modified for modern compilers.

## Requirements

You'll need `g++` compiler.

  * Debian: `apt-get install build-essential` should be enough

and run

```

$ make

```

## mult_generator

Currently does not take any parameters, all settings is hard-coded in constants:

```c

#define DIM 2        // dimensionality of the data

#define NUM 40      // number of clusters

#define MAXMU 10    // mean in each dimension is in range [0,MAXMU]

#define MINMU -10

#define MINSIGMA 0

#define MAXSIGMA 20*sqrt(DIM) // standard deviation (to be added on top

// of row sum in each dimension is in range [0,MAXSIGMA]

#define MAXSIZE 100  // size of each cluster is in range [MINSIZE,MAXSIZE]

#define MINSIZE 10

#define RUNS 10      // number of data sets to be generated

```

simply run:

```

$ ./mult_generator

```

![mult_generator](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/2d-4c-no9.png)

## elly

Ellipsoid generator

```

$ ./elly [-k ] [-d ] [-s ]

```

where all parameters are optional and:

  * `` is a positive int >= 2

  * `` is a positive int >= 2

  * `` is a long int.

![elly example](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/elly-2d10c13s.png)

## cure

CURE data sets generator. See Guha, Sudipto, Rajeev Rastogi, and Kyuseok Shim. "CURE: an efficient clustering algorithm for

large databases." ACM SIGMOD Record. Vol. 27. No. 2. ACM, 1998. for more details.

The distribution of data points is just approximated

```

$ ./cure -n  [-d ] [-s ] [-l ] [-m ] [-t type of data]

```

where:

  * `-l` minimal x/y value

  * `-m` maximal x/y value

  * `-t` type of dataset, currently supports values 0-2

![cure t0](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/cure-t0-2k-2d.png)

![cure t1](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/cure-t1-2k-2d.png)

![cure t2](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/cure-t2-4k.png)

## disk

Disk in disk dataset are two clusters formed by a circle and an annulus around it.

![disk-4600](https://raw.githubusercontent.com/deric/handl-data-generators/screens/img/disk-4600.png)

## Authors

  * Julia Handl

  * Joshua Knowles

  * John Burkardt

  * Tomas Barton