Show simple record

dc.contributor.authorZhu, Lei
dc.contributor.authorPang, Shaoning
dc.contributor.authorSarrafzadeh, Hossein
dc.contributor.authorBan, Tao
dc.contributor.authorInoue, Daisuke
dc.date.accessioned2016-10-19T21:43:55Z
dc.date.available2016-10-19T21:43:55Z
dc.date.issued2016-04-13
dc.identifier.issn1041-4347
dc.identifier.issn1558-2191
dc.identifier.urihttps://hdl.handle.net/10652/3582
dc.description.abstractMax-flow has been adopted for semi-supervised data modelling, yet existing algorithms were derived only for the learning from static data. This paper proposes an online max-flow algorithm for the semi-supervised learning from data streams. Consider a graph learned from labelled and unlabelled data, and the graph being updated dynamically for accommodating online data adding and retiring. In learning from the resulting non stationary graph, we augment and de-augment paths to update max-flow with a theoretical guarantee that the updated max-flow equals to that from batch retraining. For classification, we compute min-cut over current max-flow, so that minimized number of similar sample pairs are classified into distinct classes. Empirical evaluation on real-world data reveals that our algorithm outperforms state-of-the-art stream classification algorithms.en_NZ
dc.language.isoenen_NZ
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en_NZ
dc.rightsAll rights reserveden_NZ
dc.subjectgraph mincutsen_NZ
dc.subjectdata modellingen_NZ
dc.subjectonline semi-supervised learningen_NZ
dc.subjectmax-flowen_NZ
dc.subjectaugmenting pathen_NZ
dc.subjectincremental decremental max-flowen_NZ
dc.subjectresidual graphen_NZ
dc.subjectalgorithmsen_NZ
dc.titleIncremental and decremental max-flow for online semi-supervised learningen_NZ
dc.typeJournal Articleen_NZ
dc.rights.holderInstitute of Electrical and Electronics Engineers (IEEE)en_NZ
dc.subject.marsden080109 Pattern Recognition and Data Miningen_NZ
dc.identifier.bibliographicCitationZhu, L., Pang, S., Sarrafzadeh, A., Ban, T., & Inoue, D. (2016). Incremental and Decremental Max-flow for Online Semi-supervised Learning. IEEE Transactions on Knowledge and Data Engineering, 28, pp.1-13.en_NZ
unitec.institutionUnitec Institute of Technologyen_NZ
unitec.publication.spage1en_NZ
unitec.publication.lpage13en_NZ
unitec.publication.volume28en_NZ
unitec.publication.titleIEEE Transactions on Knowledge and Data Engineeringen_NZ
unitec.peerreviewedyesen_NZ
dc.contributor.affiliationUnitec Institute of Technologyen_NZ
dc.contributor.affiliationNational Institution of Information and Communications Technology (Tokyo, Japan)en_NZ
unitec.identifier.roms58923en_NZ
unitec.identifier.roms61219
unitec.institution.studyareaComputing


Files in this item

Thumbnail

This item appears in

Show simple record