Active and semi-supervised data domain description

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Standard

Active and semi-supervised data domain description. / Görnitz, Nico; Kloft, Marius; Brefeld, Ulf.
Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I. ed. / Wray Buntine; Marko Grobelnik; Dunja Mladenic; John Shawe-Taylor. Berlin, Heidelberg: Springer Verlag, 2009. p. 407-422 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5781 LNAI, No. PART 1).

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Harvard

Görnitz, N, Kloft, M & Brefeld, U 2009, Active and semi-supervised data domain description. in W Buntine, M Grobelnik, D Mladenic & J Shawe-Taylor (eds), Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 5781 LNAI, Springer Verlag, Berlin, Heidelberg, pp. 407-422, European Conference on Machine Learning and Knowledge Discovery in Databases - 2009, Bled, Slovenia, 07.09.09. https://doi.org/10.1007/978-3-642-04180-8_44

APA

Görnitz, N., Kloft, M., & Brefeld, U. (2009). Active and semi-supervised data domain description. In W. Buntine, M. Grobelnik, D. Mladenic, & J. Shawe-Taylor (Eds.), Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I (pp. 407-422). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5781 LNAI, No. PART 1). Springer Verlag. https://doi.org/10.1007/978-3-642-04180-8_44

Vancouver

Görnitz N, Kloft M, Brefeld U. Active and semi-supervised data domain description. In Buntine W, Grobelnik M, Mladenic D, Shawe-Taylor J, editors, Machine Learning and Knowledge Discovery in Databases: European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I. Berlin, Heidelberg: Springer Verlag. 2009. p. 407-422. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-642-04180-8_44

Bibtex

@inbook{ca1a8bdc8dc34c3f8a7703bc2b43d1f7,
title = "Active and semi-supervised data domain description",
abstract = "Data domain description techniques aim at deriving concise descriptions of objects belonging to a category of interest. For instance, the support vector domain description (SVDD) learns a hypersphere enclosing the bulk of provided unlabeled data such that points lying outside of the ball are considered anomalous. However, relevant information such as expert and background knowledge remain unused in the unsupervised setting. In this paper, we rephrase data domain description as a semi-supervised learning task, that is, we propose a semi-supervised generalization of data domain description (SSSVDD) to process unlabeled and labeled examples. The corresponding optimization problem is non-convex. We translate it into an unconstraint, continuous problem that can be optimized accurately by gradient-based techniques. Furthermore, we devise an effective active learning strategy to query low-confidence observations. Our empirical evaluation on network intrusion detection and object recognition tasks shows that our SSSVDDs consistently outperform baseline methods in relevant learning settings.",
keywords = "Informatics, Active Learning, Background knowledge, Baseline methods, Continuous problems, Data domain description, Empirical evaluations, Gradient based, Learning settings, Network intrusion detection, Optimization problems, Semi-supervised learning, upport vector domain description, Unlabeled data, Business informatics",
author = "Nico G{\"o}rnitz and Marius Kloft and Ulf Brefeld",
year = "2009",
month = jul,
day = "1",
doi = "10.1007/978-3-642-04180-8_44",
language = "English",
isbn = "978-3-642-04179-2",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
number = "PART 1",
pages = "407--422",
editor = "Wray Buntine and Marko Grobelnik and Dunja Mladenic and John Shawe-Taylor",
booktitle = "Machine Learning and Knowledge Discovery in Databases",
address = "Germany",
note = "European Conference on Machine Learning and Knowledge Discovery in Databases - 2009, ECML-PKDD ; Conference date: 07-09-2009 Through 11-09-2009",
url = "https://www.k4all.org/event/european-conference-on-machine-learning-and-principles-and-practice-of-knowledge-discovery-in-databases/",

}

RIS

TY - CHAP

T1 - Active and semi-supervised data domain description

AU - Görnitz, Nico

AU - Kloft, Marius

AU - Brefeld, Ulf

PY - 2009/7/1

Y1 - 2009/7/1

N2 - Data domain description techniques aim at deriving concise descriptions of objects belonging to a category of interest. For instance, the support vector domain description (SVDD) learns a hypersphere enclosing the bulk of provided unlabeled data such that points lying outside of the ball are considered anomalous. However, relevant information such as expert and background knowledge remain unused in the unsupervised setting. In this paper, we rephrase data domain description as a semi-supervised learning task, that is, we propose a semi-supervised generalization of data domain description (SSSVDD) to process unlabeled and labeled examples. The corresponding optimization problem is non-convex. We translate it into an unconstraint, continuous problem that can be optimized accurately by gradient-based techniques. Furthermore, we devise an effective active learning strategy to query low-confidence observations. Our empirical evaluation on network intrusion detection and object recognition tasks shows that our SSSVDDs consistently outperform baseline methods in relevant learning settings.

AB - Data domain description techniques aim at deriving concise descriptions of objects belonging to a category of interest. For instance, the support vector domain description (SVDD) learns a hypersphere enclosing the bulk of provided unlabeled data such that points lying outside of the ball are considered anomalous. However, relevant information such as expert and background knowledge remain unused in the unsupervised setting. In this paper, we rephrase data domain description as a semi-supervised learning task, that is, we propose a semi-supervised generalization of data domain description (SSSVDD) to process unlabeled and labeled examples. The corresponding optimization problem is non-convex. We translate it into an unconstraint, continuous problem that can be optimized accurately by gradient-based techniques. Furthermore, we devise an effective active learning strategy to query low-confidence observations. Our empirical evaluation on network intrusion detection and object recognition tasks shows that our SSSVDDs consistently outperform baseline methods in relevant learning settings.

KW - Informatics

KW - Active Learning

KW - Background knowledge

KW - Baseline methods

KW - Continuous problems

KW - Data domain description

KW - Empirical evaluations

KW - Gradient based

KW - Learning settings

KW - Network intrusion detection

KW - Optimization problems

KW - Semi-supervised learning

KW - upport vector domain description

KW - Unlabeled data

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=70350627210&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-04180-8_44

DO - 10.1007/978-3-642-04180-8_44

M3 - Article in conference proceedings

AN - SCOPUS:70350627210

SN - 978-3-642-04179-2

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 407

EP - 422

BT - Machine Learning and Knowledge Discovery in Databases

A2 - Buntine, Wray

A2 - Grobelnik, Marko

A2 - Mladenic, Dunja

A2 - Shawe-Taylor, John

PB - Springer Verlag

CY - Berlin, Heidelberg

T2 - European Conference on Machine Learning and Knowledge Discovery in Databases - 2009

Y2 - 7 September 2009 through 11 September 2009

ER -

Recently viewed

Activities

  1. Presentation: Nexus of Housing and Migration
  2. It's how, not what we use that matters - Communications Modes in the Internet
  3. Understanding Societal Development and Moral Progress: The Contribution of the World Values Surveys
  4. Knowledge Spaces
  5. Workshop „Meta-Image Day 2012”
  6. Liquidity, Flows, Circulation: The Cultural Logic of Environmentalization (2nd part) 2021
  7. Language Learning in Blended-Learning Projects: Moodle, Web 2.0, and Learner Agency
  8. Ars Electronica
  9. Blogs in the Foreign Language Classroom
  10. 9th International Multi-Conference on Systems, Signals and Devices - SSD 2012
  11. Are Self-Employed Time and Money Poor? Dynamics of Interpendent Multidimensional Poverty with German Time Use Diary Data
  12. Developing the ‘Benign by Design’ Approach for a Rational Design of Green Derivatives of b -Blockers: Propranolol as an Example
  13. From Christiane to Elisabeth. The 19th Century Genesis of the Intellectually Working Woman and the Epistemological Dependency on Structures of Desire in Hegel and Nietzsche
  14. Institutional dynamics of affecting and being affected: The emotionalization of injustice and the threat of withdrawing the organizational identification
  15. Scene as Ecosystem, Scenes as Parts of Ecosystems or Scene versus Ecosystem? Some considerations about the compability of two conceptional approaches
  16. 24th IEEE International Conference on Business Informatics
  17. 13th Trends in Enterprise Architecture Research Workshop - TEAR 2018
  18. Exploring Sustainability in Virtual Space
  19. Towards a sustainable Southern Transylvania: Recognizing existing contributions to reach sustainable visions and empowering stakeholders
  20. 1st International Conference of the International Association for Computing and Philosophy - IACAP 2011
  21. Going Green 2015 - Exploring Sustainability in Virtual Space

Publications

  1. Modern Baselines for SPARQL Semantic Parsing
  2. Inconsistent short-term effects of enhanced structural complexity on soil microbial properties across German forests
  3. Reporting and Analysing the Environmental Impact of Language Models on the Example of Commonsense Question Answering with External Knowledge
  4. A blueprint for mapping and modelling ecosystem services
  5. Influence of initial severity of depression on effectiveness of low intensity interventions
  6. An Experimental Approach to the Optimization of Customer Information at the Point of Sale
  7. Model Based Logistic Monitoring of Assembly Areas
  8. Automated scoring in the era of artificial intelligence
  9. Handicaps in job assignment
  10. Adaptor device for transmitting e.g. blood pressure data of human body from blood pressure measuring device of data communication system to e.g. personal computer, has controller for controlling transmission of data to communication module
  11. A common European asylum system? How variation in Member States’ administrative capacity undermines EU asylum harmonisation
  12. Basic analysis of the incremental profile forming process
  13. HAWK@QALD5 - Trying to answer hybrid questions with various simple ranking techniques
  14. Models for integrated production-inventory systems
  15. Modeling of microstructural pattern formation in crystal plasticity
  16. Learning in Real-World Laboratories: A Systematic Impulse for Discussion
  17. Time for the Environment: The Tutzing Time Ecology Project
  18. Evidence for singlet state β cleavage in the photoreaction of α-(2,6-dimethoxyphenoxy)-acetophenone inferred from time-resolved CIDNP spectroscopy
  19. Distal and proximal predictors of snacking at work
  20. A Theory-Based Concept for Fostering Sustainability Competencies in Engineering Programs