Multivariate supervised discretization, a neighborhood graph approach

Muhlenbach, Fabrice; Rakotomalala, Ricco

doi:10.1109/icdm.2002.1183918

Cited by 11 publications

(6 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The neighborhood graphs, which are special tools of the computational geometry, can be used in the clustering algorithms based on regions of influence, but also in many other data mining tasks, and especially in the supervised machine learning [17], [18], [19], [20]. Such neighborhood structure can be for example the k-nearest neighbor, the Delaunay triangulation, the MST, the relative neighborhood graph (RNG), the Gabriel graph (GG).…”

Section: Gbc Methods a Neighborhood Graphsmentioning

confidence: 99%

A New Clustering Algorithm Based on Regions of Influence with Self-Detection of the Best Number of Clusters

Muhlenbach

Lallich

2009

2009 Ninth IEEE International Conference on Data Mining

View full text Add to dashboard Cite

Abstract-Clustering methods usually require to know the best number of clusters, or another parameter, e.g. a threshold, which is not ever easy to provide. This paper proposes a new graph-based clustering method called "GBC" which detects automatically the best number of clusters, without requiring any other parameter. In this method based on regions of influence, a graph is constructed and the edges of the graph having the higher values are cut according to a hierarchical divisive procedure. An index is calculated from the size average of the cut edges which self-detects the more appropriate number of clusters. The results of GBC for 3 quality indices (Dunn, Silhouette and Davies-Bouldin) are compared with those of K-Means, Ward's hierarchical clustering method and DBSCAN on 8 benchmarks. The experiments show the good performance of GBC in the case of well separated clusters, even if the data are unbalanced, non-convex or with presence of outliers, whatever the shape of the clusters.

show abstract

Section: Gbc Methods a Neighborhood Graphsmentioning

confidence: 99%

A New Clustering Algorithm Based on Regions of Influence with Self-Detection of the Best Number of Clusters

Muhlenbach

Lallich

2009

2009 Ninth IEEE International Conference on Data Mining

View full text Add to dashboard Cite

show abstract

“…The discretization process converts continuous variables into discrete attributes (Dougherty et al 1995;Fayyad and Irani 1993;Frank and Witten 1999;Liu et al 2002;Muhlenbach and Rakotomalala 2002). This conversion constructs intervals which may be disjoint or overlapping (i.e.…”

Section: Discretizationmentioning

confidence: 99%

“…Indeed in classification fields, both methods use a discretization process but the discretization type is different. As explained in Muhlenbach and Rakotomalala (2005), discretization (Dougherty et al 1995;Zighed et al 1997;Frank and Witten 1999;Muhlenbach and Rakotomalala 2002) allows continuous data to be converted into intervals (i.e. discrete data) whose boundaries are defined using object values.…”

Section: Introductionmentioning

confidence: 99%

Dichotomic lattices and local discretization for Galois lattices

Girard¹,

Bertet²,

Visani³

2015

Adv Data Anal Classif

View full text Add to dashboard Cite

show abstract

“…The neighborhood graphs, which are special tools of the computational geometry, can be used in many data mining tasks, and especially in the supervised machine learning [15], [16], [17], [18]. Such neighborhood structure can be for example the minimum spanning tree (MST), the relative neighborhood graph (RNG) [19], the Gabriel graph (GG) or the Delaunay triangulation.…”

Section: From the Social Network To The Neighborhood Graphmentioning

confidence: 99%

Discovering Research Communities by Clustering Bibliographical Data

Muhlenbach

Lallich²

2010

2010 IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology

View full text Add to dashboard Cite

Today's world is characterized by the multiplicity of interconnections through many types of links between the people, that is why mining social networks appears to be an important topic. Extracting information from social networks becomes a challenging problem, particularly in the case of the discovery of community structures.Mining bibliographical data can be useful to find communities of researchers. In this paper we propose a formal definition to consider the similarity and dissimilarity between individuals of a social network and how a graph-based clustering method can extract research communities from the DBLP database.

show abstract

Multivariate supervised discretization, a neighborhood graph approach

Cited by 11 publications

References 2 publications

A New Clustering Algorithm Based on Regions of Influence with Self-Detection of the Best Number of Clusters

A New Clustering Algorithm Based on Regions of Influence with Self-Detection of the Best Number of Clusters

Dichotomic lattices and local discretization for Galois lattices

Discovering Research Communities by Clustering Bibliographical Data

Contact Info

Product

Resources

About