dc.contributor.author | Zhu, Lei | |
dc.contributor.author | Pang, Shaoning | |
dc.contributor.author | Sarrafzadeh, Hossein | |
dc.contributor.author | Ban, Tao | |
dc.contributor.author | Inoue, Daisuke | |
dc.date.accessioned | 2016-10-19T21:43:55Z | |
dc.date.available | 2016-10-19T21:43:55Z | |
dc.date.issued | 2016-04-13 | |
dc.identifier.issn | 1041-4347 | |
dc.identifier.issn | 1558-2191 | |
dc.identifier.uri | https://hdl.handle.net/10652/3582 | |
dc.description.abstract | Max-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.iso | en | en_NZ |
dc.publisher | Institute of Electrical and Electronics Engineers (IEEE) | en_NZ |
dc.rights | All rights reserved | en_NZ |
dc.subject | graph mincuts | en_NZ |
dc.subject | data modelling | en_NZ |
dc.subject | online semi-supervised learning | en_NZ |
dc.subject | max-flow | en_NZ |
dc.subject | augmenting path | en_NZ |
dc.subject | incremental decremental max-flow | en_NZ |
dc.subject | residual graph | en_NZ |
dc.subject | algorithms | en_NZ |
dc.title | Incremental and decremental max-flow for online semi-supervised learning | en_NZ |
dc.type | Journal Article | en_NZ |
dc.rights.holder | Institute of Electrical and Electronics Engineers (IEEE) | en_NZ |
dc.subject.marsden | 080109 Pattern Recognition and Data Mining | en_NZ |
dc.identifier.bibliographicCitation | Zhu, 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.institution | Unitec Institute of Technology | en_NZ |
unitec.publication.spage | 1 | en_NZ |
unitec.publication.lpage | 13 | en_NZ |
unitec.publication.volume | 28 | en_NZ |
unitec.publication.title | IEEE Transactions on Knowledge and Data Engineering | en_NZ |
unitec.peerreviewed | yes | en_NZ |
dc.contributor.affiliation | Unitec Institute of Technology | en_NZ |
dc.contributor.affiliation | National Institution of Information and Communications Technology (Tokyo, Japan) | en_NZ |
unitec.identifier.roms | 58923 | en_NZ |
unitec.identifier.roms | 61219 | |
unitec.institution.studyarea | Computing | |