Home > Research > Publications & Outputs > Semi-supervised Spectral Connectivity Projectio...

Links

Text available via DOI:

View graph of relations

Semi-supervised Spectral Connectivity Projection Pursuit

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Semi-supervised Spectral Connectivity Projection Pursuit. / Hofmeyr, David; Pavlidis, Nicos Georgios.
Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015 . IEEE, 2015. p. 201-206.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Hofmeyr, D & Pavlidis, NG 2015, Semi-supervised Spectral Connectivity Projection Pursuit. in Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015 . IEEE, pp. 201-206. https://doi.org/10.1109/RoboMech.2015.7359523

APA

Hofmeyr, D., & Pavlidis, N. G. (2015). Semi-supervised Spectral Connectivity Projection Pursuit. In Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015 (pp. 201-206). IEEE. https://doi.org/10.1109/RoboMech.2015.7359523

Vancouver

Hofmeyr D, Pavlidis NG. Semi-supervised Spectral Connectivity Projection Pursuit. In Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015 . IEEE. 2015. p. 201-206 doi: 10.1109/RoboMech.2015.7359523

Author

Hofmeyr, David ; Pavlidis, Nicos Georgios. / Semi-supervised Spectral Connectivity Projection Pursuit. Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015 . IEEE, 2015. pp. 201-206

Bibtex

@inproceedings{3fa172e98fb9429fa57682a5183605b2,
title = "Semi-supervised Spectral Connectivity Projection Pursuit",
abstract = "We propose a projection pursuit method based on semi-supervised spectral connectivity. The projection index is given by the second eigenvalue of the graph Laplacian of the projected data. An incomplete label set is used to modify pairwise similarities between data in such a way that penalises projections which do not admit a separation of the classes (within the training data). We show that the global optimum of the proposed problem converges to the Transductive Support Vector Machine solution, as the scaling parameter is reduced to zero. We evaluate the performance of the proposed method on benchmark data sets.",
author = "David Hofmeyr and Pavlidis, {Nicos Georgios}",
year = "2015",
month = nov,
day = "25",
doi = "10.1109/RoboMech.2015.7359523",
language = "English",
isbn = "9781467374507",
pages = "201--206",
booktitle = "Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015",
publisher = "IEEE",

}

RIS

TY - GEN

T1 - Semi-supervised Spectral Connectivity Projection Pursuit

AU - Hofmeyr, David

AU - Pavlidis, Nicos Georgios

PY - 2015/11/25

Y1 - 2015/11/25

N2 - We propose a projection pursuit method based on semi-supervised spectral connectivity. The projection index is given by the second eigenvalue of the graph Laplacian of the projected data. An incomplete label set is used to modify pairwise similarities between data in such a way that penalises projections which do not admit a separation of the classes (within the training data). We show that the global optimum of the proposed problem converges to the Transductive Support Vector Machine solution, as the scaling parameter is reduced to zero. We evaluate the performance of the proposed method on benchmark data sets.

AB - We propose a projection pursuit method based on semi-supervised spectral connectivity. The projection index is given by the second eigenvalue of the graph Laplacian of the projected data. An incomplete label set is used to modify pairwise similarities between data in such a way that penalises projections which do not admit a separation of the classes (within the training data). We show that the global optimum of the proposed problem converges to the Transductive Support Vector Machine solution, as the scaling parameter is reduced to zero. We evaluate the performance of the proposed method on benchmark data sets.

U2 - 10.1109/RoboMech.2015.7359523

DO - 10.1109/RoboMech.2015.7359523

M3 - Conference contribution/Paper

SN - 9781467374507

SP - 201

EP - 206

BT - Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 2015

PB - IEEE

ER -