Active and semi-supervised data domain description

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

Authors

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.

Original languageEnglish
Title of host publicationMachine Learning and Knowledge Discovery in Databases : European Conference, ECML PKDD 2009, Bled, Slovenia, September 7-11, 2009, Proceedings, Part I
EditorsWray Buntine, Marko Grobelnik, Dunja Mladenic, John Shawe-Taylor
Number of pages16
Place of PublicationBerlin, Heidelberg
PublisherSpringer Verlag
Publication date01.07.2009
Pages407-422
ISBN (print)978-3-642-04179-2
ISBN (electronic)978-3-642-04180-8
DOIs
Publication statusPublished - 01.07.2009
Externally publishedYes
EventEuropean Conference on Machine Learning and Knowledge Discovery in Databases - 2009 - Bled, Slovenia
Duration: 07.09.200911.09.2009
https://www.k4all.org/event/european-conference-on-machine-learning-and-principles-and-practice-of-knowledge-discovery-in-databases/

    Research areas

  • 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

Recently viewed

Publications

  1. Mechanical characterization of as-cast AA7075/6060 and CuSn6/Cu99.5 compounds using an experimental and numerical push-out test
  2. Understanding Context Collapse for Social Media Users
  3. Multifractality Versus (Mono-) Fractality as Evidence of Nonlinear Interactions Across Timescales
  4. Canopy structure influences arthropod communities within and beyond tree identity effects
  5. The dynamics of prior entry in serial visual processing
  6. Recruitment practices in small and medium size enterprises.
  7. Editorial: Effects of the Introduction of the Statutory Minimum Wage in Germany
  8. Efficacy of a web-based intervention with and without guidance for employees with risky drinking
  9. Forest structure and heterogeneity increase diversity and alter composition of host–parasitoid networks
  10. Unusual deactivation in the asymmetric hydrogenation of itaconic acid
  11. Insights into adoption of farming practices through multiple lenses
  12. Horizontal, but not vertical canopy structure is related to stand functional diversity in a subtropical slope forest
  13. How can Environmental Management contribute to Shareholder Value?
  14. Introduction lectures in entrepreneurship
  15. What do we know about new venture investment time patterns?
  16. A trainable object finder, selector and identifier for pollen, spores and other things
  17. Open-flow mixing and transfer operators
  18. Geometrical Characterization of Polyethylene Oxide Nanofibers by Atom Force Microscope and Confocal Laser Scanning Microscope
  19. "Learning by doing"
  20. “You’re Not Allowed to Give Us the Solution, but Can You Guide Us towards It?”
  21. The importance of product lifetime labelling for purchase decisions
  22. Minimal conditions of motor inductions of approach-avoidance states
  23. Analysis of brittle layer forming mechanism in Ti6Al4V sloping structures by SLM technology
  24. Audio Video Sampler
  25. Systemanalyse für Softwaresysteme
  26. The causal effects of exports on firm size and labor productivity: first evidence from a matching approach
  27. Functions of Constitutions
  28. Hot deformation behavior and processing map of Mg-3Sn-2Ca-0.4Al-0.4Zn alloy
  29. Comparative effectiveness of three versions of a stepped care model for insomnia differing in the amount of therapist support in internet-delivered treatment
  30. „Ist das dein Handy oder vibrierst du?“
  31. Political discourse as mediated and public discourse
  32. Paar normal oder paranormal
  33. A new magnesium alloy system
  34. An Analysis of Methane Mitigation as a Response to Climate Change
  35. Credit constraints, endogenous innovations, and price setting in international trade
  36. Abiotic and biotic drivers of tree trait effects on soil microbial biomass and soil carbon concentration
  37. “Greedy Buyers, Amoral Speculators and Lacking State Control”
  38. A Social–Ecological Systems Framework as a Tool for Understanding the Effectiveness of Biosphere Reserve Management
  39. Neue Rechte und Universität
  40. Bierbrausieb
  41. Is the reverse J-shaped diameter distribution universally applicable in European virgin beech forests?
  42. The effect of industrialization and globalization on domestic land-use
  43. Habitat models for the four-fingered skink (Carlia tetradactyla) at the microhabitat and landscape scale
  44. Experimental-numerical study of laser-shock-peening-induced retardation of fatigue crack propagation in Ti-17 titanium alloy
  45. Begriff und Merkmale junger Unternehmen
  46. Who are we and who are you? The strategic use of forms of address in political interviews
  47. Subsidies for learning in renewable energy technologies under market power and emission trading
  48. Best-Practice-Beispiel: Wie kann Mentoring in die neue Studienorganisation implementiert werden?