A label noise filtering and label missing supplement framework based on game theory

Simple item page
dc.citation.issue4
dc.citation.volume9
dc.contributor.authorLiu Y
dc.contributor.authorYao R
dc.contributor.authorJia S
dc.contributor.authorWang F
dc.contributor.authorWang R
dc.contributor.authorMa R
dc.contributor.authorQi L
dc.date.accessioned2023-11-27T22:13:33Z
dc.date.accessioned2024-07-25T06:42:02Z
dc.date.available2022-01-04
dc.date.available2023-11-27T22:13:33Z
dc.date.available2024-07-25T06:42:02Z
dc.date.issued2023-08-31
dc.description.abstractLabeled data is widely used in various classification tasks. However, there is a huge challenge that labels are often added artificially. Wrong labels added by malicious users will affect the training effect of the model. The unreliability of labeled data has hindered the research. In order to solve the above problems, we propose a framework of Label Noise Filtering and Missing Label Supplement (LNFS). And we take location labels in Location-Based Social Networks (LBSN) as an example to implement our framework. For the problem of label noise filtering, we first use FastText to transform the restaurant's labels into vectors, and then based on the assumption that the label most similar to all other labels in the location is most representative. We use cosine similarity to judge and select the most representative label. For the problem of label missing, we use simple common word similarity to judge the similarity of users' comments, and then use the label of the similar restaurant to supplement the missing labels. To optimize the performance of the model, we introduce game theory into our model to simulate the game between the malicious users and the model to improve the reliability of the model. Finally, a case study is given to illustrate the effectiveness and reliability of LNFS.
dc.description.confidentialfalse
dc.edition.editionAugust 2023
dc.format.pagination887-895
dc.identifier.citationLiu Y, Yao R, Jia S, Wang F, Wang R, Ma R, Qi L. (2023). A label noise filtering and label missing supplement framework based on game theory. Digital Communications and Networks. 9. 4. (pp. 887-895).
dc.identifier.doi10.1016/j.dcan.2021.12.008
dc.identifier.eissn2352-8648
dc.identifier.elements-typejournal-article
dc.identifier.issn2468-5925
dc.identifier.piiS2352864821001115
dc.identifier.urihttps://mro.massey.ac.nz/handle/10179/70708
dc.languageEnglish
dc.publisherElsevier B.V. on behalf of KeAi Communications Co Ltd for the Chongqing University of Posts and Telecommunications
dc.publisher.urihttps://www.sciencedirect.com/science/article/pii/S2352864821001115
dc.relation.isPartOfDigital Communications and Networks
dc.rightsCC BY 4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectLabel noise
dc.subjectFastText
dc.subjectCosine similarity
dc.subjectGame theory
dc.subjectLSTM
dc.titleA label noise filtering and label missing supplement framework based on game theory
dc.typeJournal article
pubs.elements-id453553
pubs.organisational-groupOther
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Published version
Size:
1.73 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Journal Articles