Matches in ScholarlyData for { <https://w3id.org/scholarlydata/inproceedings/eswc2014/paper/research/37> ?p ?o. }
Showing items 1 to 15 of
15
with 100 items per page.
- 37 creator bei-xu.
- 37 creator huakang-li.
- 37 creator mingyi-guo.
- 37 creator yi-liu.
- 37 type InProceedings.
- 37 label "A Knowledge Based Approach for Tackling Mislabeled Multi-class Big Social Data".
- 37 sameAs 37.
- 37 abstract "The performance of classification models extremely relies on the quality of training data. However, label imperfection is an inherent fault of training data, which cannot be manually handled in big data environment. Various methods have been proposed to remove label noises in order to improve classification quality, with the side effect of cutting down data bulk. In this paper, we propose a knowledge based approach for tackling mislabeled multi-class big data, in which knowledge graph technique is combined with other data correction method to perceive and correct the error labels in big data. Experiments on a medical Q&A social data set show our knowledge graph based approach can effectively improve data quality and classification accuracy. Furthermore, this approach can be applied in other data mining tasks requiring deep understanding.".
- 37 hasAuthorList authorList.
- 37 isPartOf proceedings.
- 37 keyword "classification".
- 37 keyword "knowledge graph".
- 37 keyword "label correction".
- 37 keyword "label imperfection".
- 37 title "A Knowledge Based Approach for Tackling Mislabeled Multi-class Big Social Data".