Big data scaling through metric mapping

exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We present new findings in regard to data analysis in very high dimensional spaces. We use dimensionalities up to around one million. A particular benefit of Correspondence Analysis is its suitability for carrying out an orthonormal mapping, or scaling, of power law distributed data. Power law distributed data are found in many domains. Correspondence factor analysis provides a latent semantic or principal axes mapping. Our experiments use data from digital chemistry and finance, and other statistically generated data.

Original languageEnglish
Title of host publicationData Science
EditorsFrancesco Palumbo, Angela Montanari, Maurizio Vichi
Pages295-306
Number of pages12
Edition195089
ISBN (Electronic)9783319557236
DOIs
Publication statusPublished - 5 Jul 2017
Externally publishedYes
Event15th Conference of the International Federation of Classification Societies - Bologna, Italy
Duration: 6 Jul 20158 Jul 2015
Conference number: 15
https://studylib.net/doc/10711915/ifcs-2015-call-for-papers-conference-of-the-international... (Link to Call for Papers)

Publication series

NameStudies in Classification, Data Analysis, and Knowledge Organization
ISSN (Print)1431-8814

Conference

Conference15th Conference of the International Federation of Classification Societies
Abbreviated titleIFCS 2015
CountryItaly
CityBologna
Period6/07/158/07/15
Internet address

Fingerprint

Correspondence Analysis
Simplicity
High-dimensional
Scaling
Metric
Factor analysis
Finance
Power Law
Semantics
Orthonormal
Factor Analysis
Chemistry
Dimensionality
Data analysis
Experiments
Big data
Correspondence analysis
Power law
Experiment

Cite this

Murtagh, F. (2017). Big data scaling through metric mapping: exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis. In F. Palumbo, A. Montanari, & M. Vichi (Eds.), Data Science (195089 ed., pp. 295-306). (Studies in Classification, Data Analysis, and Knowledge Organization). https://doi.org/10.1007/978-3-319-55723-6_23
Murtagh, Fionn. / Big data scaling through metric mapping : exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis. Data Science. editor / Francesco Palumbo ; Angela Montanari ; Maurizio Vichi. 195089. ed. 2017. pp. 295-306 (Studies in Classification, Data Analysis, and Knowledge Organization).
@inproceedings{a58e306ea16646bba9fb26f79a2ba6e9,
title = "Big data scaling through metric mapping: exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis",
abstract = "We present new findings in regard to data analysis in very high dimensional spaces. We use dimensionalities up to around one million. A particular benefit of Correspondence Analysis is its suitability for carrying out an orthonormal mapping, or scaling, of power law distributed data. Power law distributed data are found in many domains. Correspondence factor analysis provides a latent semantic or principal axes mapping. Our experiments use data from digital chemistry and finance, and other statistically generated data.",
author = "Fionn Murtagh",
year = "2017",
month = "7",
day = "5",
doi = "10.1007/978-3-319-55723-6_23",
language = "English",
isbn = "9783319557229",
series = "Studies in Classification, Data Analysis, and Knowledge Organization",
pages = "295--306",
editor = "Francesco Palumbo and Angela Montanari and Maurizio Vichi",
booktitle = "Data Science",
edition = "195089",

}

Murtagh, F 2017, Big data scaling through metric mapping: exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis. in F Palumbo, A Montanari & M Vichi (eds), Data Science. 195089 edn, Studies in Classification, Data Analysis, and Knowledge Organization, pp. 295-306, 15th Conference of the International Federation of Classification Societies, Bologna, Italy, 6/07/15. https://doi.org/10.1007/978-3-319-55723-6_23

Big data scaling through metric mapping : exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis. / Murtagh, Fionn.

Data Science. ed. / Francesco Palumbo; Angela Montanari; Maurizio Vichi. 195089. ed. 2017. p. 295-306 (Studies in Classification, Data Analysis, and Knowledge Organization).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Big data scaling through metric mapping

T2 - exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis

AU - Murtagh, Fionn

PY - 2017/7/5

Y1 - 2017/7/5

N2 - We present new findings in regard to data analysis in very high dimensional spaces. We use dimensionalities up to around one million. A particular benefit of Correspondence Analysis is its suitability for carrying out an orthonormal mapping, or scaling, of power law distributed data. Power law distributed data are found in many domains. Correspondence factor analysis provides a latent semantic or principal axes mapping. Our experiments use data from digital chemistry and finance, and other statistically generated data.

AB - We present new findings in regard to data analysis in very high dimensional spaces. We use dimensionalities up to around one million. A particular benefit of Correspondence Analysis is its suitability for carrying out an orthonormal mapping, or scaling, of power law distributed data. Power law distributed data are found in many domains. Correspondence factor analysis provides a latent semantic or principal axes mapping. Our experiments use data from digital chemistry and finance, and other statistically generated data.

UR - http://www.scopus.com/inward/record.url?scp=85048191403&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-55723-6_23

DO - 10.1007/978-3-319-55723-6_23

M3 - Conference contribution

SN - 9783319557229

T3 - Studies in Classification, Data Analysis, and Knowledge Organization

SP - 295

EP - 306

BT - Data Science

A2 - Palumbo, Francesco

A2 - Montanari, Angela

A2 - Vichi, Maurizio

ER -

Murtagh F. Big data scaling through metric mapping: exploiting the remarkable simplicity of very high dimensional spaces using correspondence analysis. In Palumbo F, Montanari A, Vichi M, editors, Data Science. 195089 ed. 2017. p. 295-306. (Studies in Classification, Data Analysis, and Knowledge Organization). https://doi.org/10.1007/978-3-319-55723-6_23