Learning from partially annotated sequences

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Learning from partially annotated sequences. / Fernandes, Eraldo R.; Brefeld, Ulf.
Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings. Hrsg. / Dimitrios Gunopulos; Thomas Hofmann; Donato Malerba; Michalis Vazirgiannis. PART 1. Aufl. Heidelberg, Berlin: Springer, 2011. S. 407-422 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 6911 LNAI, Nr. PART 1).

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Fernandes, ER & Brefeld, U 2011, Learning from partially annotated sequences. in D Gunopulos, T Hofmann, D Malerba & M Vazirgiannis (Hrsg.), Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings. PART 1 Aufl., Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Nr. PART 1, Bd. 6911 LNAI, Springer, Heidelberg, Berlin, S. 407-422, European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - ECML PKDD 2011, Athen, Griechenland, 05.09.11. https://doi.org/10.1007/978-3-642-23780-5_36

APA

Fernandes, E. R., & Brefeld, U. (2011). Learning from partially annotated sequences. In D. Gunopulos, T. Hofmann, D. Malerba, & M. Vazirgiannis (Hrsg.), Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings (PART 1 Aufl., S. 407-422). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Band 6911 LNAI, Nr. PART 1). Springer. https://doi.org/10.1007/978-3-642-23780-5_36

Vancouver

Fernandes ER, Brefeld U. Learning from partially annotated sequences. in Gunopulos D, Hofmann T, Malerba D, Vazirgiannis M, Hrsg., Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings. PART 1 Aufl. Heidelberg, Berlin: Springer. 2011. S. 407-422. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-642-23780-5_36

Bibtex

@inbook{0b7856608ef7418f8057a3ab0347cc36,
title = "Learning from partially annotated sequences",
abstract = "We study sequential prediction models in cases where only fragments of the sequences are annotated with the ground-truth. The task does not match the standard semi-supervised setting and is highly relevant in areas such as natural language processing, where completely labeled instances are expensive and require editorial data. We propose to generalize the semi-supervised setting and devise a simple transductive loss-augmented perceptron to learn from inexpensive partially annotated sequences that could for instance be provided by laymen, the wisdom of the crowd, or even automatically. Experiments on mono- and cross-lingual named entity recognition tasks with automatically generated partially annotated sentences from Wikipedia demonstrate the effectiveness of the proposed approach. Our results show that learning from partially labeled data is never worse than standard supervised and semi-supervised approaches trained on data with the same ratio of labeled and unlabeled tokens.",
keywords = "Informatics, Automatically generated, Cross-lingual, Labeled data, Named entity recognition, NAtural language processing, Perceptron, Semi-supervised, Sequential prediction, Hide Markov Model, Unlabeled Data, Neural Information Processing System, Entity Recognition, Annotate Sequence, Business informatics",
author = "Fernandes, {Eraldo R.} and Ulf Brefeld",
year = "2011",
doi = "10.1007/978-3-642-23780-5_36",
language = "English",
isbn = "978-3-642-23779-9",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
number = "PART 1",
pages = "407--422",
editor = "Dimitrios Gunopulos and Thomas Hofmann and Donato Malerba and Michalis Vazirgiannis",
booktitle = "Machine Learning and Knowledge Discovery in Databases - European Conference, ECML PKDD 2011, Proceedings",
address = "Germany",
edition = "PART 1",
note = "European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - ECML PKDD 2011, ECML PKDD 2011 ; Conference date: 05-09-2011 Through 09-09-2011",
url = "http://www.ecmlpkdd2011.org/, https://www.ecmlpkdd2011.org/",

}

RIS

TY - CHAP

T1 - Learning from partially annotated sequences

AU - Fernandes, Eraldo R.

AU - Brefeld, Ulf

PY - 2011

Y1 - 2011

N2 - We study sequential prediction models in cases where only fragments of the sequences are annotated with the ground-truth. The task does not match the standard semi-supervised setting and is highly relevant in areas such as natural language processing, where completely labeled instances are expensive and require editorial data. We propose to generalize the semi-supervised setting and devise a simple transductive loss-augmented perceptron to learn from inexpensive partially annotated sequences that could for instance be provided by laymen, the wisdom of the crowd, or even automatically. Experiments on mono- and cross-lingual named entity recognition tasks with automatically generated partially annotated sentences from Wikipedia demonstrate the effectiveness of the proposed approach. Our results show that learning from partially labeled data is never worse than standard supervised and semi-supervised approaches trained on data with the same ratio of labeled and unlabeled tokens.

AB - We study sequential prediction models in cases where only fragments of the sequences are annotated with the ground-truth. The task does not match the standard semi-supervised setting and is highly relevant in areas such as natural language processing, where completely labeled instances are expensive and require editorial data. We propose to generalize the semi-supervised setting and devise a simple transductive loss-augmented perceptron to learn from inexpensive partially annotated sequences that could for instance be provided by laymen, the wisdom of the crowd, or even automatically. Experiments on mono- and cross-lingual named entity recognition tasks with automatically generated partially annotated sentences from Wikipedia demonstrate the effectiveness of the proposed approach. Our results show that learning from partially labeled data is never worse than standard supervised and semi-supervised approaches trained on data with the same ratio of labeled and unlabeled tokens.

KW - Informatics

KW - Automatically generated

KW - Cross-lingual

KW - Labeled data

KW - Named entity recognition

KW - NAtural language processing

KW - Perceptron

KW - Semi-supervised

KW - Sequential prediction

KW - Hide Markov Model

KW - Unlabeled Data

KW - Neural Information Processing System

KW - Entity Recognition

KW - Annotate Sequence

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=80052421057&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/047857db-fd1a-3b48-8b27-a1acc478a333/

U2 - 10.1007/978-3-642-23780-5_36

DO - 10.1007/978-3-642-23780-5_36

M3 - Article in conference proceedings

AN - SCOPUS:80052421057

SN - 978-3-642-23779-9

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 - European Conference, ECML PKDD 2011, Proceedings

A2 - Gunopulos, Dimitrios

A2 - Hofmann, Thomas

A2 - Malerba, Donato

A2 - Vazirgiannis, Michalis

PB - Springer

CY - Heidelberg, Berlin

T2 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases - ECML PKDD 2011

Y2 - 5 September 2011 through 9 September 2011

ER -

DOI

Zuletzt angesehen

Publikationen

  1. Lengthscale-dependent modelling of ductile failure in metallic microstructures
  2. A tale of scale: Plot but not neighbourhood tree diversity increases leaf litter ant diversity
  3. Modelling of a servo piezo mechanical hydraulic actuator and its feed-forward control
  4. Motion-decoupled internal force control in grasping with visco-elastic contacts
  5. Conception of an indirectly controlled servo valve for IC engine valve position control
  6. The Lotka-Volterra Model for Competition Controlled by a Sliding Mode Approach
  7. Impact assessment of emissions stabilization scenarios with and without induced technological change
  8. Simulation of composite hot extrusion with high reinforcing Volumes
  9. Online Network Impedance Identification with Wave-Package and Inter-Harmonic Signals
  10. The relationship between values and knowledge in visioning for landscape management
  11. Bifurcation loads of beams of glued-laminated timber with intermediate lateral supports
  12. Wie lang sollte eine Rollstuhlrampe sein?
  13. The differential effects of self-view in virtual meetings when speaking vs. listening
  14. Noninteracting force/motion control of defective manipulation systems
  15. Comparing marginal effects between different models and/or samples
  16. A Unified Contextual Bandit Framework for Long- and Short-Term Recommendations
  17. A switching observer for sensorless control of an electromagnetic valve actuator for camless internal combustion engines
  18. From stories to maps
  19. Sense of Place in Spatial Planning
  20. Solution for the direct kinematics problem of the general stewart-gough platform by using only linear actuators’ orientations
  21. CETUS – a baseline approach to type extraction
  22. Imitation and interindividual differences
  23. Frame-based Optimal Design
  24. Control of geometry deviation by stiffness variation in polymer deep drawing tools
  25. Comparative study of resonant circuit for power transmission via inductive link
  26. Non-invasive approaches for phenotyping of enhanced performance traits in bean
  27. Investigation of the utilization of oat pomace and acid whey in technical scale succinic acid fermentation including downstream processing
  28. Assessing mire-specific biodiversity with an indicator based approach
  29. The interaction of ecological norm orientation and external factors in the domain of travel mode choice behavior
  30. Facilitating collaborative processes in transdisciplinary research using design prototyping
  31. 3D characterization of beta-phases in AZ91D by synchrotron-radiation based microtomography
  32. Experimental and numerical analysis of material flow in porthole die extrusion
  33. Testing for Economies of Scope in European Railways
  34. Bifurcation loads of circular curved beams of glued-laminated timber with continuous lateral support
  35. Effects of different video- or text-based reflection stimuli on pre-service teachers’ emotions, immersion, cognitive load and knowledge-based reasoning
  36. Cognitive performance limitations in operating rooms
  37. Current and New Research Perspectives on Dynamic Facial Emotion Detection in Emotional Interface
  38. Consumer response to monochrome Guideline Daily Amount nutrition labels
  39. Non-destructive transmissive inductive thickness sensor for IoT applications
  40. Functional trait similarity of native and invasive herb species in subtropical China-Environment-specific differences are the key
  41. In situ synchrotron diffraction of the solidification of Mg-RE alloys
  42. MICSIM: Concept, Developments, and Applications of a PC Microsimulation Model for Research and Teaching
  43. UAV remote laser scanner improvement by continuous scanning using DC motors
  44. Decoding evidence-based entrepreneurship
  45. Microstructure and corrosion of AZ91 with small amounts of cerium
  46. Tree species and functional traits but not species richness affect interrill erosion processes in young subtropical forests