Advanced search
1 file | 580.19 KB Add to list

Dealing with data veracity in multiple criteria handling : an LSP-based sibling approach

Author
Organization
Project
Abstract
In a big data context, data often originate from various unreliable sources and cannot be considered perfect. Data veracity denotes the overall confidence we have in the data and clearly has an impact on the results of querying and decision making processes. In this paper, we study the impact of data veracity on criterion handling and propose a novel, LSP-based sibling evaluation approach that explicitly copes with data veracity. Logic Scoring of Preference (LSP) is a computational intelligence method that is based on logic criteria selection, evaluation and aggregation. In our proposal, LSP techniques are independently used for scoring preferences on data and preferences on data confidence. This results for each preference on data in an elementary sibling pair, consisting of a satisfaction score and its associated confidence score. Sibling pairs are aggregated using a novel sibling aggregation structure. The resulting sibling pairs, being indicators of both suitability and confidence, provide better interpretable and explainable evaluation results.
Keywords
Big data, Veracity, Querying, Decision support, Interpretable computational intelligence, LSP, DATA QUALITY, INFORMATION, FUZZY

Downloads

  • (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 580.19 KB

Citation

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

MLA
De Tré, Guy, and Jozo J. Dujmovic. “Dealing with Data Veracity in Multiple Criteria Handling : An LSP-Based Sibling Approach.” FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021), edited by T. Andreasen et al., vol. 12871, Springer, 2021, pp. 82–96, doi:10.1007/978-3-030-86967-0_7.
APA
De Tré, G., & Dujmovic, J. J. (2021). Dealing with data veracity in multiple criteria handling : an LSP-based sibling approach. In T. Andreasen, G. De Tré, J. Kacprzyk, H. L. Larsen, G. Bordogna, & S. Zadrozny (Eds.), FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021) (Vol. 12871, pp. 82–96). https://doi.org/10.1007/978-3-030-86967-0_7
Chicago author-date
De Tré, Guy, and Jozo J. Dujmovic. 2021. “Dealing with Data Veracity in Multiple Criteria Handling : An LSP-Based Sibling Approach.” In FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021), edited by T. Andreasen, Guy De Tré, J. Kacprzyk, H. L. Larsen, G. Bordogna, and S. Zadrozny, 12871:82–96. Springer. https://doi.org/10.1007/978-3-030-86967-0_7.
Chicago author-date (all authors)
De Tré, Guy, and Jozo J. Dujmovic. 2021. “Dealing with Data Veracity in Multiple Criteria Handling : An LSP-Based Sibling Approach.” In FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021), ed by. T. Andreasen, Guy De Tré, J. Kacprzyk, H. L. Larsen, G. Bordogna, and S. Zadrozny, 12871:82–96. Springer. doi:10.1007/978-3-030-86967-0_7.
Vancouver
1.
De Tré G, Dujmovic JJ. Dealing with data veracity in multiple criteria handling : an LSP-based sibling approach. In: Andreasen T, De Tré G, Kacprzyk J, Larsen HL, Bordogna G, Zadrozny S, editors. FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021). Springer; 2021. p. 82–96.
IEEE
[1]
G. De Tré and J. J. Dujmovic, “Dealing with data veracity in multiple criteria handling : an LSP-based sibling approach,” in FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021), Bratislava, SLOVAKIA, 2021, vol. 12871, pp. 82–96.
@inproceedings{8731302,
  abstract     = {{In a big data context, data often originate from various unreliable sources and cannot be considered perfect. Data veracity denotes the overall confidence we have in the data and clearly has an impact on the results of querying and decision making processes. In this paper, we study the impact of data veracity on criterion handling and propose a novel, LSP-based sibling evaluation approach that explicitly copes with data veracity. Logic Scoring of Preference (LSP) is a computational intelligence method that is based on logic criteria selection, evaluation and aggregation. In our proposal, LSP techniques are independently used for scoring preferences on data and preferences on data confidence. This results for each preference on data in an elementary sibling pair, consisting of a satisfaction score and its associated confidence score. Sibling pairs are aggregated using a novel sibling aggregation structure. The resulting sibling pairs, being indicators of both suitability and confidence, provide better interpretable and explainable evaluation results.}},
  author       = {{De Tré, Guy and Dujmovic, Jozo J.}},
  booktitle    = {{FLEXIBLE QUERY ANSWERING SYSTEMS (FQAS 2021)}},
  editor       = {{Andreasen, T. and De Tré, Guy and Kacprzyk, J. and Larsen, H. L. and Bordogna, G. and Zadrozny, S.}},
  isbn         = {{9783030869663}},
  issn         = {{0302-9743}},
  keywords     = {{Big data,Veracity,Querying,Decision support,Interpretable computational intelligence,LSP,DATA QUALITY,INFORMATION,FUZZY}},
  language     = {{eng}},
  location     = {{Bratislava, SLOVAKIA}},
  pages        = {{82--96}},
  publisher    = {{Springer}},
  title        = {{Dealing with data veracity in multiple criteria handling : an LSP-based sibling approach}},
  url          = {{http://doi.org/10.1007/978-3-030-86967-0_7}},
  volume       = {{12871}},
  year         = {{2021}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: