### Abstract

We propose a generative model based on a Delaunay graph to learn the topology of a set of points. It uses the maximum likelihood principle to tune its parameters. This work is a first step towards a topological model of a set of points grounded on statistics.

Original language | English |
---|---|

Title of host publication | WSOM 2005 - 5th Workshop on Self-Organizing Maps |

Pages | 347-354 |

Number of pages | 8 |

Publication status | Published - 2005 |

Externally published | Yes |

Event | 5th Workshop on Self-Organizing Maps, WSOM 2005 - Paris Duration: 5 Sep 2005 → 8 Sep 2005 |

### Other

Other | 5th Workshop on Self-Organizing Maps, WSOM 2005 |
---|---|

City | Paris |

Period | 5/9/05 → 8/9/05 |

### Fingerprint

### Keywords

- Delaunay graph
- Generative model
- Maximum likelihood
- Mixture model
- Topology modelling

### ASJC Scopus subject areas

- Information Systems

### Cite this

*WSOM 2005 - 5th Workshop on Self-Organizing Maps*(pp. 347-354)

**A generative Gaussian graph to learn the topology of a set of points.** / Aupetit, Michael.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

*WSOM 2005 - 5th Workshop on Self-Organizing Maps.*pp. 347-354, 5th Workshop on Self-Organizing Maps, WSOM 2005, Paris, 5/9/05.

}

TY - GEN

T1 - A generative Gaussian graph to learn the topology of a set of points

AU - Aupetit, Michael

PY - 2005

Y1 - 2005

N2 - We propose a generative model based on a Delaunay graph to learn the topology of a set of points. It uses the maximum likelihood principle to tune its parameters. This work is a first step towards a topological model of a set of points grounded on statistics.

AB - We propose a generative model based on a Delaunay graph to learn the topology of a set of points. It uses the maximum likelihood principle to tune its parameters. This work is a first step towards a topological model of a set of points grounded on statistics.

KW - Delaunay graph

KW - Generative model

KW - Maximum likelihood

KW - Mixture model

KW - Topology modelling

UR - http://www.scopus.com/inward/record.url?scp=84890285778&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890285778&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84890285778

SP - 347

EP - 354

BT - WSOM 2005 - 5th Workshop on Self-Organizing Maps

ER -