Advanced search
1 file | 174.28 KB Add to list
Author
Organization
Abstract
Several authors have developed relation extraction methods for automatically learning or refining taxonomies from large text corpora such as the Web. However, without appropriate post-processing, such taxonomies are often inconsistent (e.g. they contain cycles). A standard approach to repairing such inconsistencies is to identify a minimally consistent subset of the extracted facts. For example, we could aim to minimize the sum of the confidence weights of the facts that are removed for restoring consistency. In this paper, we present MAP inference as a base method for this approach, and analyze how it can be improved by taking into account dependencies between the extracted facts. These dependencies correspond to rules of thumb such as “if a given fact is wrong then all facts that have been extracted from the same sentence are also likely to be wrong", which we encode in Markov logic. We present experimental results to demonstrate the potential of this idea.
Keywords
MAP inference, Markov Logic, inconsistency, Taxonomy extraction

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 174.28 KB

Citation

Please use this url to cite or link to this publication:

MLA
Merhej, Elie, Steven Schockaert, Martine De Cock, et al. “Repairing Inconsistent Taxonomies Using MAP Inference and Rules of Thumb.” Web-KR  ’14 : Proceedings of the 5th International Workshop on Web-scale Knowledge Representation Retrieval and Reasoning. New York, NY, USA: Association for Computing Machinery (ACM), 2014. 31–36. Print.
APA
Merhej, E., Schockaert, S., De Cock, M., Blondeel, M., Alfarone, D., & Davis, J. (2014). Repairing inconsistent taxonomies using MAP inference and rules of thumb. Web-KR  ’14 : proceedings of the 5th international workshop on web-scale knowledge representation retrieval and reasoning (pp. 31–36). Presented at the 5th International workshop on Web-scale Knowledge Representation Retrieval & Reasoning (Web-KR 2014), New York, NY, USA: Association for Computing Machinery (ACM).
Chicago author-date
Merhej, Elie, Steven Schockaert, Martine De Cock, Marjon Blondeel, Daniele Alfarone, and Jesse Davis. 2014. “Repairing Inconsistent Taxonomies Using MAP Inference and Rules of Thumb.” In Web-KR  ’14 : Proceedings of the 5th International Workshop on Web-scale Knowledge Representation Retrieval and Reasoning, 31–36. New York, NY, USA: Association for Computing Machinery (ACM).
Chicago author-date (all authors)
Merhej, Elie, Steven Schockaert, Martine De Cock, Marjon Blondeel, Daniele Alfarone, and Jesse Davis. 2014. “Repairing Inconsistent Taxonomies Using MAP Inference and Rules of Thumb.” In Web-KR  ’14 : Proceedings of the 5th International Workshop on Web-scale Knowledge Representation Retrieval and Reasoning, 31–36. New York, NY, USA: Association for Computing Machinery (ACM).
Vancouver
1.
Merhej E, Schockaert S, De Cock M, Blondeel M, Alfarone D, Davis J. Repairing inconsistent taxonomies using MAP inference and rules of thumb. Web-KR  ’14 : proceedings of the 5th international workshop on web-scale knowledge representation retrieval and reasoning. New York, NY, USA: Association for Computing Machinery (ACM); 2014. p. 31–6.
IEEE
[1]
E. Merhej, S. Schockaert, M. De Cock, M. Blondeel, D. Alfarone, and J. Davis, “Repairing inconsistent taxonomies using MAP inference and rules of thumb,” in Web-KR ’14 : proceedings of the 5th international workshop on web-scale knowledge representation retrieval and reasoning, Shanghai, PR China, 2014, pp. 31–36.
@inproceedings{5757779,
  abstract     = {Several authors have developed relation extraction methods for automatically learning or refining taxonomies from large text corpora such as the Web. However, without appropriate post-processing, such taxonomies are often inconsistent (e.g. they contain cycles). A standard approach to repairing such inconsistencies is to identify a minimally consistent subset of the extracted facts. For example, we could aim to minimize the sum of the confidence weights of the facts that are removed for restoring consistency. In this paper, we present MAP inference as a base method for this approach, and analyze how it can be improved by taking into account dependencies between the extracted facts. These dependencies correspond to rules of thumb such as “if a given fact is wrong then all facts that have been extracted from the same sentence are also likely to be wrong", which we encode in Markov logic. We present experimental results to demonstrate the potential of this idea.},
  author       = {Merhej, Elie and Schockaert, Steven and De Cock, Martine and Blondeel, Marjon and Alfarone, Daniele and Davis, Jesse},
  booktitle    = {Web-KR '14 : proceedings of the 5th international workshop on web-scale knowledge representation retrieval and reasoning},
  isbn         = {9781450316064},
  keywords     = {MAP inference,Markov Logic,inconsistency,Taxonomy extraction},
  language     = {eng},
  location     = {Shanghai, PR China},
  pages        = {31--36},
  publisher    = {Association for Computing Machinery (ACM)},
  title        = {Repairing inconsistent taxonomies using MAP inference and rules of thumb},
  url          = {http://dx.doi.org/10.1145/2663792.2663804},
  year         = {2014},
}

Altmetric
View in Altmetric