Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/eswc2014/paper/research/82> ?p ?o. }
Showing items 1 to 13 of
13
with 100 items per page.
- 82 creator axel-cyrille-ngonga-ngomo.
- 82 creator klaus-lyko.
- 82 creator mohamed-sherif.
- 82 type InProceedings.
- 82 label "Unsupervised Link Discovery Through Knowledge Base Repair".
- 82 sameAs 82.
- 82 abstract "Over the last few years, devising efficient approaches to compute links between datasets has been regarded as central to achieve the vision behind the Data Web. Several unsupervised approaches have been developed to achieve this goal. Yet, so far, none of these approaches make use of replication of resources across several knowledge bases. In this paper, we present Colibri, an unsupervised approach that allows discovering links between datasets while improving the quality of the instance data in these datasets. A Colibri iteration begins by generating links between knowledge bases. Then, our approach makes use of these links to detect resources with probably erroneous or missing information. The erroneous or missing information detected by the approach is finally corrected or added. We evaluate our approach on benchmark datasets with respect to the F-score it achieves. Our results suggest that Colibri can improve the results of unsupervised machine-learning approaches for link discovery by up to 12% while correctly detecting and repairing erroneous resources.".
- 82 hasAuthorList authorList.
- 82 isPartOf proceedings.
- 82 keyword "Instance Repair".
- 82 keyword "Link Discovery".
- 82 keyword "Unsupervised Machine Learning".
- 82 title "Unsupervised Link Discovery Through Knowledge Base Repair".