Statistical tools to assess the reliability of self-organizing maps

Eric De Bodt; Marie Cottrell; Michel Verleysen

hal-00122768, version 1

Statistical tools to assess the reliability of self-organizing maps

Eric De Bodt ^1, 2, Marie Cottrell () ^3, 4, Michel Verleysen () ⁵

Neural Networks 15, n° 8-9 (2002) 967-978

Abstract: Results of neural network learning are always subject to some variability, due to the sensitivity to initial conditions, to convergence to local minima, and, sometimes more dramatically, to sampling variability. This paper presents a set of tools designed to assess the reliability of the results of Self-Organizing Maps (SOM), i.e. to test on a statistical basis the confidence we can have on the result of a specific SOM. The tools concern the quantization error in a SOM, and the neighborhood relations (both at the level of a specific pair of observations and globally on the map). As a by-product, these measures also allow to assess the adequacy of the number of units chosen in a map. The tools may also be used to measure objectively how the SOM are less sensitive to non-linear optimization problems (local minima, convergence, etc.) than other neural network models.

1: ESA (ESA)
Université Lille II - Droit et santé
2: IAG-FIN (IAG-FIN)
Université Catholique de Louvain (UCL) - Belgique
3: Statistique Appliquée et MOdélisation Stochastique (SAMOS)
Université Paris I - Panthéon-Sorbonne
4: Modélisation Appliquée, Trajectoires Institutionnelles et Stratégies Socio-Économiques (MATISSE)
CNRS : UMR8595 – Université Paris I - Panthéon-Sorbonne
5: MachineLearning Group - DICE (DICE)
Université Catholique de Louvain (UCL) - Belgique

hal-00122768, version 1
http://hal.archives-ouvertes.fr/hal-00122768
oai:hal.archives-ouvertes.fr:hal-00122768
From: Marie Cottrell <>
Submitted on: Thursday, 4 January 2007 16:57:58
Updated on: Monday, 22 January 2007 15:21:39

See detailed view

Associated documents

PDF :

arXiv: math.ST/0701144

Export

Bibtex EndNote TEI RefWorks

<?xml version="1.0" encoding="UTF-8"?>
<listBibl xmlns="http://www.tei-c.org/ns/1.0" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML">
<listBibl type="article"><head>Articles in peer-reviewed journal</head>
<biblStruct type="article" xml:id="hal-00122768"><analytic>
<author><persName><forename>Eric</forename><surname>De Bodt</surname></persName>
<affiliation>
<orgName type="laboratory">ESA</orgName>
<orgName type="team">ESA</orgName>
<country>FR</country>
<orgName type="institution">Université Lille II - Droit et santé</orgName>
</affiliation>
<affiliation>
<orgName type="laboratory">IAG-FIN</orgName>
<orgName type="team">IAG-FIN</orgName>
<country>BE</country>
<orgName type="institution">Université Catholique de Louvain (UCL) - Belgique</orgName>
</affiliation>
</author>
<author><persName><forename>Marie</forename><surname>Cottrell</surname></persName>
<affiliation>
<orgName type="laboratory">SAMOS</orgName>
<orgName type="team">Statistique Appliquée et MOdélisation Stochastique</orgName>
<country>FR</country>
<orgName type="institution">Université Paris I - Panthéon-Sorbonne</orgName>
</affiliation>
<affiliation>
<orgName type="laboratory">MATISSE</orgName>
<orgName type="team">Modélisation Appliquée, Trajectoires Institutionnelles et Stratégies Socio-Économiques</orgName>
<country>FR</country>
<orgName type="institution">CNRS : UMR8595</orgName>
<orgName type="institution">Université Paris I - Panthéon-Sorbonne</orgName>
</affiliation>
</author>
<author><persName><forename>Michel</forename><surname>Verleysen</surname></persName>
<affiliation>
<orgName type="laboratory">DICE</orgName>
<orgName type="team">MachineLearning Group - DICE</orgName>
<country>BE</country>
<orgName type="institution">Université Catholique de Louvain (UCL) - Belgique</orgName>
</affiliation>
</author>
<title type="main" level="a">Statistical tools to assess the reliability of self-organizing maps</title></analytic><monogr><title level="j" type="main">Neural Networks</title><imprint><biblScope type="publicationDate">2002-00-00</biblScope><biblScope type="vol">15, n° 8-9</biblScope><biblScope type="pp">967-978</biblScope></imprint></monogr><idno type="HALid">http://hal.archives-ouvertes.fr/hal-00122768</idno><idno type="HALFile">http://hal.archives-ouvertes.fr/hal-00122768/PDF/paper_DeBodt_revised.pdf</idno></biblStruct>
</listBibl>
</listBibl>