Knowledge Discovery from Complex High Dimensional Data

Sangkyun Lee, Andreas Holzinger

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

Modern data analysis is confronted by increasing dimensionality of problems, mainly contributed by higher resolutions available for data acquisition and by our use of larger models with more degrees of freedom to investigate complex systems deeper. High dimensionality constitutes one aspect of “big data”, which brings us not only computational but also statistical and perceptional challenges. Most data analysis problems are solved using techniques of optimization, where large-scale optimization requires faster algorithms and implementations. Computed solutions must be evaluated for statistical quality, since otherwise false discoveries can be made. Recent papers suggest to control and modify algorithms themselves for better statistical properties. Finally, human perception puts an inherent limit on our understanding to three dimensional spaces, making it almost impossible to grasp complex phenomena. For aid, we use dimensionality reduction or other techniques, but these usually do not capture relations between interesting objects. Here graph-based knowledge representation has lots of potential, for instance to create perceivable and interactive representations and to perform new types of analysis based on graph theory and network topology. In this
article, we show glimpses of new developments in these aspects.
LanguageEnglish
Title of host publicationSolving Large Scale Learning Tasks. Challenges and Algorithms
Subtitle of host publicationSpringer Lecture Notes in Artificial Intelligence LNAI 9580
EditorsStefan Michaelis, Nico Piatkowski, Marco Stolpe
Place of PublicationHeidelberg, Berlin, New York
PublisherSpringer International
Pages148
Number of pages167
ISBN (Electronic)978-3-319-41706-6
ISBN (Print)978-3-319-41705-9
DOIs
StatusPublished - 10 Jul 2016
EventWorkshop Machine Learning for Biomedicine at TU Graz - TU Graz, Graz, Austria
Duration: 26 Jan 201626 Jan 2016

Conference

ConferenceWorkshop Machine Learning for Biomedicine at TU Graz
CountryAustria
CityGraz
Period26/01/1626/01/16

Fingerprint

Data mining
Graph theory
Knowledge representation
Large scale systems
Data acquisition
Topology
Big data

Keywords

  • Machine Learning
  • Health Informatics
  • Graph Learning
  • Graph-Based Data Mining

ASJC Scopus subject areas

  • Artificial Intelligence

Fields of Expertise

  • Information, Communication & Computing

Treatment code (Nähere Zuordnung)

  • Basic - Fundamental (Grundlagenforschung)

Cite this

Lee, S., & Holzinger, A. (2016). Knowledge Discovery from Complex High Dimensional Data. In S. Michaelis, N. Piatkowski, & M. Stolpe (Eds.), Solving Large Scale Learning Tasks. Challenges and Algorithms: Springer Lecture Notes in Artificial Intelligence LNAI 9580 (pp. 148). Heidelberg, Berlin, New York: Springer International. DOI: 10.1007/978-3-319-41706-6_7

Knowledge Discovery from Complex High Dimensional Data. / Lee, Sangkyun; Holzinger, Andreas.

Solving Large Scale Learning Tasks. Challenges and Algorithms: Springer Lecture Notes in Artificial Intelligence LNAI 9580. ed. / Stefan Michaelis; Nico Piatkowski; Marco Stolpe. Heidelberg, Berlin, New York : Springer International, 2016. p. 148.

Research output: Chapter in Book/Report/Conference proceedingChapter

Lee, S & Holzinger, A 2016, Knowledge Discovery from Complex High Dimensional Data. in S Michaelis, N Piatkowski & M Stolpe (eds), Solving Large Scale Learning Tasks. Challenges and Algorithms: Springer Lecture Notes in Artificial Intelligence LNAI 9580. Springer International, Heidelberg, Berlin, New York, pp. 148, Workshop Machine Learning for Biomedicine at TU Graz, Graz, Austria, 26/01/16. DOI: 10.1007/978-3-319-41706-6_7
Lee S, Holzinger A. Knowledge Discovery from Complex High Dimensional Data. In Michaelis S, Piatkowski N, Stolpe M, editors, Solving Large Scale Learning Tasks. Challenges and Algorithms: Springer Lecture Notes in Artificial Intelligence LNAI 9580. Heidelberg, Berlin, New York: Springer International. 2016. p. 148. Available from, DOI: 10.1007/978-3-319-41706-6_7
Lee, Sangkyun ; Holzinger, Andreas. / Knowledge Discovery from Complex High Dimensional Data. Solving Large Scale Learning Tasks. Challenges and Algorithms: Springer Lecture Notes in Artificial Intelligence LNAI 9580. editor / Stefan Michaelis ; Nico Piatkowski ; Marco Stolpe. Heidelberg, Berlin, New York : Springer International, 2016. pp. 148
@inbook{79001e9ba33f44289f3857d94d3c4abd,
title = "Knowledge Discovery from Complex High Dimensional Data",
abstract = "Modern data analysis is confronted by increasing dimensionality of problems, mainly contributed by higher resolutions available for data acquisition and by our use of larger models with more degrees of freedom to investigate complex systems deeper. High dimensionality constitutes one aspect of “big data”, which brings us not only computational but also statistical and perceptional challenges. Most data analysis problems are solved using techniques of optimization, where large-scale optimization requires faster algorithms and implementations. Computed solutions must be evaluated for statistical quality, since otherwise false discoveries can be made. Recent papers suggest to control and modify algorithms themselves for better statistical properties. Finally, human perception puts an inherent limit on our understanding to three dimensional spaces, making it almost impossible to grasp complex phenomena. For aid, we use dimensionality reduction or other techniques, but these usually do not capture relations between interesting objects. Here graph-based knowledge representation has lots of potential, for instance to create perceivable and interactive representations and to perform new types of analysis based on graph theory and network topology. In thisarticle, we show glimpses of new developments in these aspects.",
keywords = "Machine Learning, Health Informatics, Graph Learning, Graph-Based Data Mining",
author = "Sangkyun Lee and Andreas Holzinger",
year = "2016",
month = "7",
day = "10",
doi = "10.1007/978-3-319-41706-6_7",
language = "English",
isbn = "978-3-319-41705-9",
pages = "148",
editor = "Stefan Michaelis and Nico Piatkowski and Marco Stolpe",
booktitle = "Solving Large Scale Learning Tasks. Challenges and Algorithms",
publisher = "Springer International",

}

TY - CHAP

T1 - Knowledge Discovery from Complex High Dimensional Data

AU - Lee,Sangkyun

AU - Holzinger,Andreas

PY - 2016/7/10

Y1 - 2016/7/10

N2 - Modern data analysis is confronted by increasing dimensionality of problems, mainly contributed by higher resolutions available for data acquisition and by our use of larger models with more degrees of freedom to investigate complex systems deeper. High dimensionality constitutes one aspect of “big data”, which brings us not only computational but also statistical and perceptional challenges. Most data analysis problems are solved using techniques of optimization, where large-scale optimization requires faster algorithms and implementations. Computed solutions must be evaluated for statistical quality, since otherwise false discoveries can be made. Recent papers suggest to control and modify algorithms themselves for better statistical properties. Finally, human perception puts an inherent limit on our understanding to three dimensional spaces, making it almost impossible to grasp complex phenomena. For aid, we use dimensionality reduction or other techniques, but these usually do not capture relations between interesting objects. Here graph-based knowledge representation has lots of potential, for instance to create perceivable and interactive representations and to perform new types of analysis based on graph theory and network topology. In thisarticle, we show glimpses of new developments in these aspects.

AB - Modern data analysis is confronted by increasing dimensionality of problems, mainly contributed by higher resolutions available for data acquisition and by our use of larger models with more degrees of freedom to investigate complex systems deeper. High dimensionality constitutes one aspect of “big data”, which brings us not only computational but also statistical and perceptional challenges. Most data analysis problems are solved using techniques of optimization, where large-scale optimization requires faster algorithms and implementations. Computed solutions must be evaluated for statistical quality, since otherwise false discoveries can be made. Recent papers suggest to control and modify algorithms themselves for better statistical properties. Finally, human perception puts an inherent limit on our understanding to three dimensional spaces, making it almost impossible to grasp complex phenomena. For aid, we use dimensionality reduction or other techniques, but these usually do not capture relations between interesting objects. Here graph-based knowledge representation has lots of potential, for instance to create perceivable and interactive representations and to perform new types of analysis based on graph theory and network topology. In thisarticle, we show glimpses of new developments in these aspects.

KW - Machine Learning

KW - Health Informatics

KW - Graph Learning

KW - Graph-Based Data Mining

U2 - 10.1007/978-3-319-41706-6_7

DO - 10.1007/978-3-319-41706-6_7

M3 - Chapter

SN - 978-3-319-41705-9

SP - 148

BT - Solving Large Scale Learning Tasks. Challenges and Algorithms

PB - Springer International

CY - Heidelberg, Berlin, New York

ER -