Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity
Abstract Identifying structure underlying high-dimensional data is a common challenge across scientific disciplines. We revisit correspondence analysis (CA), a classical method revealing such structures, from a network perspective. We present the poorly-known equivalence of CA to spectral clustering...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Publishing Group
2021-04-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-87971-9 |
| id |
doaj-4fb9c06fecf040749cdc0863644aa633 |
|---|---|
| record_format |
Article |
| spelling |
doaj-4fb9c06fecf040749cdc0863644aa6332021-05-02T11:37:36ZengNature Publishing GroupScientific Reports2045-23222021-04-0111111410.1038/s41598-021-87971-9Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexityAlje van Dam0Mark Dekker1Ignacio Morales-Castilla2Miguel Á. Rodríguez3David Wichmann4Mara Baudena5Copernicus Institute of Sustainable Development, Utrecht UniversityCentre for Complex Systems Studies, Utrecht UniversityGloCEE-Global Change Ecology and Evolution Group, Department of Life Sciences, University of AlcaláGloCEE-Global Change Ecology and Evolution Group, Department of Life Sciences, University of AlcaláCentre for Complex Systems Studies, Utrecht UniversityCopernicus Institute of Sustainable Development, Utrecht UniversityAbstract Identifying structure underlying high-dimensional data is a common challenge across scientific disciplines. We revisit correspondence analysis (CA), a classical method revealing such structures, from a network perspective. We present the poorly-known equivalence of CA to spectral clustering and graph-embedding techniques. We point out a number of complementary interpretations of CA results, other than its traditional interpretation as an ordination technique. These interpretations relate to the structure of the underlying networks. We then discuss an empirical example drawn from ecology, where we apply CA to the global distribution of Carnivora species to show how both the clustering and ordination interpretation can be used to find gradients in clustered data. In the second empirical example, we revisit the economic complexity index as an application of correspondence analysis, and use the different interpretations of the method to shed new light on the empirical results within this literature.https://doi.org/10.1038/s41598-021-87971-9 |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Alje van Dam Mark Dekker Ignacio Morales-Castilla Miguel Á. Rodríguez David Wichmann Mara Baudena |
| spellingShingle |
Alje van Dam Mark Dekker Ignacio Morales-Castilla Miguel Á. Rodríguez David Wichmann Mara Baudena Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity Scientific Reports |
| author_facet |
Alje van Dam Mark Dekker Ignacio Morales-Castilla Miguel Á. Rodríguez David Wichmann Mara Baudena |
| author_sort |
Alje van Dam |
| title |
Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| title_short |
Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| title_full |
Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| title_fullStr |
Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| title_full_unstemmed |
Correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| title_sort |
correspondence analysis, spectral clustering and graph embedding: applications to ecology and economic complexity |
| publisher |
Nature Publishing Group |
| series |
Scientific Reports |
| issn |
2045-2322 |
| publishDate |
2021-04-01 |
| description |
Abstract Identifying structure underlying high-dimensional data is a common challenge across scientific disciplines. We revisit correspondence analysis (CA), a classical method revealing such structures, from a network perspective. We present the poorly-known equivalence of CA to spectral clustering and graph-embedding techniques. We point out a number of complementary interpretations of CA results, other than its traditional interpretation as an ordination technique. These interpretations relate to the structure of the underlying networks. We then discuss an empirical example drawn from ecology, where we apply CA to the global distribution of Carnivora species to show how both the clustering and ordination interpretation can be used to find gradients in clustered data. In the second empirical example, we revisit the economic complexity index as an application of correspondence analysis, and use the different interpretations of the method to shed new light on the empirical results within this literature. |
| url |
https://doi.org/10.1038/s41598-021-87971-9 |
| work_keys_str_mv |
AT aljevandam correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity AT markdekker correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity AT ignaciomoralescastilla correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity AT miguelarodriguez correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity AT davidwichmann correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity AT marabaudena correspondenceanalysisspectralclusteringandgraphembeddingapplicationstoecologyandeconomiccomplexity |
| _version_ |
1721491837718364160 |