Model selection for Gaussian latent block clustering with the integrated classification likelihood View Full Text


Ontology type: schema:ScholarlyArticle     


Article Info

DATE

2014-02-09

AUTHORS

Aurore Lomet, Gérard Govaert, Yves Grandvalet

ABSTRACT

Block clustering aims to reveal homogeneous block structures in a data table. Among the different approaches of block clustering, we consider here a model-based method: the Gaussian latent block model for continuous data which is an extension of the Gaussian mixture model for one-way clustering. For a given data table, several candidate models are usually examined, which differ for example in the number of clusters. Model selection then becomes a critical issue. To this end, we develop a criterion based on an approximation of the integrated classification likelihood for the Gaussian latent block model, and propose a Bayesian information criterion-like variant following the same pattern. We also propose a non-asymptotic exact criterion, thus circumventing the controversial definition of the asymptotic regime arising from the dual nature of the rows and columns in co-clustering. The experimental results show steady performances of these criteria for medium to large data tables. More... »

PAGES

489-508

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/s11634-013-0161-3

DOI

http://dx.doi.org/10.1007/s11634-013-0161-3

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1015532627


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/01", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Mathematical Sciences", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0104", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Statistics", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France", 
          "id": "http://www.grid.ac/institutes/grid.462261.5", 
          "name": [
            "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Lomet", 
        "givenName": "Aurore", 
        "id": "sg:person.010044306034.33", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010044306034.33"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France", 
          "id": "http://www.grid.ac/institutes/grid.462261.5", 
          "name": [
            "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Govaert", 
        "givenName": "G\u00e9rard", 
        "id": "sg:person.013354526745.06", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013354526745.06"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France", 
          "id": "http://www.grid.ac/institutes/grid.462261.5", 
          "name": [
            "Heudiasyc, Universit\u00e9 de Technologie de Compi\u00e8gne, CNRS UMR 7253, Compi\u00e8gne, France"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Grandvalet", 
        "givenName": "Yves", 
        "id": "sg:person.015255215731.52", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015255215731.52"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/s11222-011-9233-4", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1045038483", 
          "https://doi.org/10.1007/s11222-011-9233-4"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s00357-008-9005-9", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1026044676", 
          "https://doi.org/10.1007/s00357-008-9005-9"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s003570000003", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1053390302", 
          "https://doi.org/10.1007/s003570000003"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1186/1471-2105-8-s10-s5", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1006792021", 
          "https://doi.org/10.1186/1471-2105-8-s10-s5"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11222-014-9472-2", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1007854802", 
          "https://doi.org/10.1007/s11222-014-9472-2"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/s11222-007-9046-7", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1000777117", 
          "https://doi.org/10.1007/s11222-007-9046-7"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2014-02-09", 
    "datePublishedReg": "2014-02-09", 
    "description": "Block clustering aims to reveal homogeneous block structures in a data table. Among the different approaches of block clustering, we consider here a model-based method: the Gaussian latent block model for continuous data which is an extension of the Gaussian mixture model for one-way clustering. For a given data table, several candidate models are usually examined, which differ for example in the number of clusters. Model selection then becomes a critical issue. To this end, we develop a criterion based on an approximation of the integrated classification likelihood for the Gaussian latent block model, and propose a Bayesian information criterion-like variant following the same pattern. We also propose a non-asymptotic exact criterion, thus circumventing the controversial definition of the asymptotic regime arising from the dual nature of the rows and columns in co-clustering. The experimental results show steady performances of these criteria for medium to large data tables.", 
    "genre": "article", 
    "id": "sg:pub.10.1007/s11634-013-0161-3", 
    "inLanguage": "en", 
    "isAccessibleForFree": false, 
    "isPartOf": [
      {
        "id": "sg:journal.1045303", 
        "issn": [
          "1862-5347", 
          "1862-5355"
        ], 
        "name": "Advances in Data Analysis and Classification", 
        "publisher": "Springer Nature", 
        "type": "Periodical"
      }, 
      {
        "issueNumber": "3", 
        "type": "PublicationIssue"
      }, 
      {
        "type": "PublicationVolume", 
        "volumeNumber": "12"
      }
    ], 
    "keywords": [
      "latent block model", 
      "integrated classification likelihood", 
      "block clustering", 
      "data tables", 
      "classification likelihood", 
      "one-way clustering", 
      "large data tables", 
      "Gaussian mixture model", 
      "number of clusters", 
      "model-based method", 
      "block model", 
      "model selection", 
      "mixture model", 
      "clustering", 
      "experimental results", 
      "critical issue", 
      "steady performance", 
      "block structure", 
      "candidate models", 
      "different approaches", 
      "table", 
      "continuous data", 
      "model", 
      "selection", 
      "performance", 
      "issues", 
      "extension", 
      "asymptotic regime", 
      "example", 
      "clusters", 
      "block", 
      "definition", 
      "method", 
      "exact criteria", 
      "criteria", 
      "data", 
      "number", 
      "approximation", 
      "rows", 
      "end", 
      "results", 
      "dual nature", 
      "variants", 
      "likelihood", 
      "nature", 
      "structure", 
      "patterns", 
      "controversial definition", 
      "medium", 
      "approach", 
      "regime", 
      "same pattern", 
      "column", 
      "homogeneous block structures", 
      "Gaussian latent block model", 
      "Bayesian information criterion-like variant", 
      "information criterion-like variant", 
      "criterion-like variant", 
      "non-asymptotic exact criterion", 
      "Gaussian latent block", 
      "latent block"
    ], 
    "name": "Model selection for Gaussian latent block clustering with the integrated classification likelihood", 
    "pagination": "489-508", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1015532627"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/s11634-013-0161-3"
        ]
      }
    ], 
    "sameAs": [
      "https://doi.org/10.1007/s11634-013-0161-3", 
      "https://app.dimensions.ai/details/publication/pub.1015532627"
    ], 
    "sdDataset": "articles", 
    "sdDatePublished": "2022-01-01T18:33", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-springernature-scigraph/baseset/20220101/entities/gbq_results/article/article_624.jsonl", 
    "type": "ScholarlyArticle", 
    "url": "https://doi.org/10.1007/s11634-013-0161-3"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s11634-013-0161-3'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s11634-013-0161-3'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s11634-013-0161-3'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s11634-013-0161-3'


 

This table displays all metadata directly associated to this object as RDF triples.

157 TRIPLES      22 PREDICATES      92 URIs      78 LITERALS      6 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/s11634-013-0161-3 schema:about anzsrc-for:01
2 anzsrc-for:0104
3 schema:author N3380eee2504f4a0d942e6a07a4965bbf
4 schema:citation sg:pub.10.1007/s00357-008-9005-9
5 sg:pub.10.1007/s003570000003
6 sg:pub.10.1007/s11222-007-9046-7
7 sg:pub.10.1007/s11222-011-9233-4
8 sg:pub.10.1007/s11222-014-9472-2
9 sg:pub.10.1186/1471-2105-8-s10-s5
10 schema:datePublished 2014-02-09
11 schema:datePublishedReg 2014-02-09
12 schema:description Block clustering aims to reveal homogeneous block structures in a data table. Among the different approaches of block clustering, we consider here a model-based method: the Gaussian latent block model for continuous data which is an extension of the Gaussian mixture model for one-way clustering. For a given data table, several candidate models are usually examined, which differ for example in the number of clusters. Model selection then becomes a critical issue. To this end, we develop a criterion based on an approximation of the integrated classification likelihood for the Gaussian latent block model, and propose a Bayesian information criterion-like variant following the same pattern. We also propose a non-asymptotic exact criterion, thus circumventing the controversial definition of the asymptotic regime arising from the dual nature of the rows and columns in co-clustering. The experimental results show steady performances of these criteria for medium to large data tables.
13 schema:genre article
14 schema:inLanguage en
15 schema:isAccessibleForFree false
16 schema:isPartOf N4c8f32cc5a4048a391dcd72e41cc1aa8
17 Nd4981add7515422391d7a85f31a9ed35
18 sg:journal.1045303
19 schema:keywords Bayesian information criterion-like variant
20 Gaussian latent block
21 Gaussian latent block model
22 Gaussian mixture model
23 approach
24 approximation
25 asymptotic regime
26 block
27 block clustering
28 block model
29 block structure
30 candidate models
31 classification likelihood
32 clustering
33 clusters
34 column
35 continuous data
36 controversial definition
37 criteria
38 criterion-like variant
39 critical issue
40 data
41 data tables
42 definition
43 different approaches
44 dual nature
45 end
46 exact criteria
47 example
48 experimental results
49 extension
50 homogeneous block structures
51 information criterion-like variant
52 integrated classification likelihood
53 issues
54 large data tables
55 latent block
56 latent block model
57 likelihood
58 medium
59 method
60 mixture model
61 model
62 model selection
63 model-based method
64 nature
65 non-asymptotic exact criterion
66 number
67 number of clusters
68 one-way clustering
69 patterns
70 performance
71 regime
72 results
73 rows
74 same pattern
75 selection
76 steady performance
77 structure
78 table
79 variants
80 schema:name Model selection for Gaussian latent block clustering with the integrated classification likelihood
81 schema:pagination 489-508
82 schema:productId N1889796c9bf7483699ea2739957eeb40
83 Na51abcf1bd784e188cc10e7d615924c4
84 schema:sameAs https://app.dimensions.ai/details/publication/pub.1015532627
85 https://doi.org/10.1007/s11634-013-0161-3
86 schema:sdDatePublished 2022-01-01T18:33
87 schema:sdLicense https://scigraph.springernature.com/explorer/license/
88 schema:sdPublisher Nec49af5af93f4011a159635488cbadd2
89 schema:url https://doi.org/10.1007/s11634-013-0161-3
90 sgo:license sg:explorer/license/
91 sgo:sdDataset articles
92 rdf:type schema:ScholarlyArticle
93 N01e18166b6794e5290f424d10ab07b6b rdf:first sg:person.013354526745.06
94 rdf:rest N2807f42408d5405a9f203ee5805a8d7f
95 N1889796c9bf7483699ea2739957eeb40 schema:name dimensions_id
96 schema:value pub.1015532627
97 rdf:type schema:PropertyValue
98 N2807f42408d5405a9f203ee5805a8d7f rdf:first sg:person.015255215731.52
99 rdf:rest rdf:nil
100 N3380eee2504f4a0d942e6a07a4965bbf rdf:first sg:person.010044306034.33
101 rdf:rest N01e18166b6794e5290f424d10ab07b6b
102 N4c8f32cc5a4048a391dcd72e41cc1aa8 schema:volumeNumber 12
103 rdf:type schema:PublicationVolume
104 Na51abcf1bd784e188cc10e7d615924c4 schema:name doi
105 schema:value 10.1007/s11634-013-0161-3
106 rdf:type schema:PropertyValue
107 Nd4981add7515422391d7a85f31a9ed35 schema:issueNumber 3
108 rdf:type schema:PublicationIssue
109 Nec49af5af93f4011a159635488cbadd2 schema:name Springer Nature - SN SciGraph project
110 rdf:type schema:Organization
111 anzsrc-for:01 schema:inDefinedTermSet anzsrc-for:
112 schema:name Mathematical Sciences
113 rdf:type schema:DefinedTerm
114 anzsrc-for:0104 schema:inDefinedTermSet anzsrc-for:
115 schema:name Statistics
116 rdf:type schema:DefinedTerm
117 sg:journal.1045303 schema:issn 1862-5347
118 1862-5355
119 schema:name Advances in Data Analysis and Classification
120 schema:publisher Springer Nature
121 rdf:type schema:Periodical
122 sg:person.010044306034.33 schema:affiliation grid-institutes:grid.462261.5
123 schema:familyName Lomet
124 schema:givenName Aurore
125 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.010044306034.33
126 rdf:type schema:Person
127 sg:person.013354526745.06 schema:affiliation grid-institutes:grid.462261.5
128 schema:familyName Govaert
129 schema:givenName Gérard
130 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.013354526745.06
131 rdf:type schema:Person
132 sg:person.015255215731.52 schema:affiliation grid-institutes:grid.462261.5
133 schema:familyName Grandvalet
134 schema:givenName Yves
135 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015255215731.52
136 rdf:type schema:Person
137 sg:pub.10.1007/s00357-008-9005-9 schema:sameAs https://app.dimensions.ai/details/publication/pub.1026044676
138 https://doi.org/10.1007/s00357-008-9005-9
139 rdf:type schema:CreativeWork
140 sg:pub.10.1007/s003570000003 schema:sameAs https://app.dimensions.ai/details/publication/pub.1053390302
141 https://doi.org/10.1007/s003570000003
142 rdf:type schema:CreativeWork
143 sg:pub.10.1007/s11222-007-9046-7 schema:sameAs https://app.dimensions.ai/details/publication/pub.1000777117
144 https://doi.org/10.1007/s11222-007-9046-7
145 rdf:type schema:CreativeWork
146 sg:pub.10.1007/s11222-011-9233-4 schema:sameAs https://app.dimensions.ai/details/publication/pub.1045038483
147 https://doi.org/10.1007/s11222-011-9233-4
148 rdf:type schema:CreativeWork
149 sg:pub.10.1007/s11222-014-9472-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1007854802
150 https://doi.org/10.1007/s11222-014-9472-2
151 rdf:type schema:CreativeWork
152 sg:pub.10.1186/1471-2105-8-s10-s5 schema:sameAs https://app.dimensions.ai/details/publication/pub.1006792021
153 https://doi.org/10.1186/1471-2105-8-s10-s5
154 rdf:type schema:CreativeWork
155 grid-institutes:grid.462261.5 schema:alternateName Heudiasyc, Université de Technologie de Compiègne, CNRS UMR 7253, Compiègne, France
156 schema:name Heudiasyc, Université de Technologie de Compiègne, CNRS UMR 7253, Compiègne, France
157 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...