Adaptive edge weighting for graph-based learning algorithms

 |  Login

Show simple item record

dc.contributor Aalto-yliopisto fi
dc.contributor Aalto University en
dc.contributor.author Karasuyama, Masayuki
dc.contributor.author Mamitsuka, Hiroshi
dc.date.accessioned 2017-05-11T08:36:17Z
dc.date.available 2017-05-11T08:36:17Z
dc.date.issued 2017
dc.identifier.citation Karasuyama , M & Mamitsuka , H 2017 , ' Adaptive edge weighting for graph-based learning algorithms ' Machine Learning , vol 106 , no. 2 , pp. 307-335 . DOI: 10.1007/s10994-016-5607-3 en
dc.identifier.issn 0885-6125
dc.identifier.issn 1573-0565
dc.identifier.other PURE UUID: 885b0463-e8a9-45f7-babe-97182fae4202
dc.identifier.other PURE ITEMURL: https://research.aalto.fi/en/publications/adaptive-edge-weighting-for-graphbased-learning-algorithms(885b0463-e8a9-45f7-babe-97182fae4202).html
dc.identifier.other PURE LINK: http://www.scopus.com/inward/record.url?scp=84995770408&partnerID=8YFLogxK
dc.identifier.other PURE FILEURL: https://research.aalto.fi/files/11519875/art_10.1007_s10994_016_5607_3.pdf
dc.identifier.uri https://aaltodoc.aalto.fi/handle/123456789/25671
dc.description.abstract Graph-based learning algorithms including label propagation and spectral clustering are known as the effective state-of-the-art algorithms for a variety of tasks in machine learning applications. Given input data, i.e. feature vectors, graph-based methods typically proceed with the following three steps: (1) generating graph edges, (2) estimating edge weights and (3) running a graph based algorithm. The first and second steps are difficult, especially when there are only a few (or no) labeled instances, while they are important because the performance of graph-based methods heavily depends on the quality of the input graph. For the second step of the three-step procedure, we propose a new method, which optimizes edge weights through a local linear reconstruction error minimization under a constraint that edges are parameterized by a similarity function of node pairs. As a result our generated graph can capture the manifold structure of the input data, where each edge represents similarity of each node pair. To further justify this approach, we also provide analytical considerations for our formulation such as an interpretation as a cross-validation of a propagation model in the feature space, and an error analysis based on a low dimensional manifold model. Experimental results demonstrated the effectiveness of our adaptive edge weighting strategy both in synthetic and real datasets. en
dc.format.extent 307-335
dc.format.mimetype application/pdf
dc.language.iso en en
dc.relation.ispartofseries Machine Learning en
dc.relation.ispartofseries Volume 106, issue 2 en
dc.rights openAccess en
dc.subject.other Software en
dc.subject.other Artificial Intelligence en
dc.subject.other 113 Computer and information sciences en
dc.title Adaptive edge weighting for graph-based learning algorithms en
dc.type A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä fi
dc.description.version Peer reviewed en
dc.contributor.department Nagoya Institute of Technology
dc.contributor.department Department of Computer Science
dc.subject.keyword Clustering
dc.subject.keyword Edge weighting
dc.subject.keyword Graph-based learning
dc.subject.keyword Manifold assumption
dc.subject.keyword Semi-supervised learning
dc.subject.keyword Software
dc.subject.keyword Artificial Intelligence
dc.subject.keyword 113 Computer and information sciences
dc.identifier.urn URN:NBN:fi:aalto-201705114055
dc.identifier.doi 10.1007/s10994-016-5607-3
dc.type.version publishedVersion


Files in this item

Files Size Format View

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record

Search archive


Advanced Search

article-iconSubmit a publication

Browse

My Account