Locality Preserving Feature Learning

Quanquan Gu, Marina Danilevsky, Zhenhui Li, Jiawei Han
; Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:477-485, 2012.

Abstract

Locality Preserving Indexing (LPI) has been quite successful in tackling document analysis problems, such as clustering or classification. The approach relies on the Locality Preserving Criterion, which preserves the locality of the data points. However, LPI takes every word in a data corpus into account, even though many words may not be useful for document clustering. To overcome this problem, we propose an approach called Locality Preserving Feature Learning (LPFL), which incorporates feature selection into LPI. Specifically, we aim to find a subset of features, and learn a linear transformation to optimize the Locality Preserving Criterion based on these features. The resulting optimization problem is a mixed integer programming problem, which we relax into a constrained Frobenius norm minimization problem, and solve using a variation of Alternating Direction Method (ADM). ADM, which iteratively updates the linear transformation matrix, the residue matrix and the Lagrangian multiplier, is theoretically guaranteed to converge at the rate O(1/t). Experiments on benchmark document datasets show that our proposed method outperforms LPI, as well as other state-of-the-art document analysis approaches.

Cite this Paper


BibTeX
@InProceedings{pmlr-v22-gu12, title = {Locality Preserving Feature Learning}, author = {Quanquan Gu and Marina Danilevsky and Zhenhui Li and Jiawei Han}, booktitle = {Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics}, pages = {477--485}, year = {2012}, editor = {Neil D. Lawrence and Mark Girolami}, volume = {22}, series = {Proceedings of Machine Learning Research}, address = {La Palma, Canary Islands}, month = {21--23 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v22/gu12/gu12.pdf}, url = {http://proceedings.mlr.press/v22/gu12.html}, abstract = {Locality Preserving Indexing (LPI) has been quite successful in tackling document analysis problems, such as clustering or classification. The approach relies on the Locality Preserving Criterion, which preserves the locality of the data points. However, LPI takes every word in a data corpus into account, even though many words may not be useful for document clustering. To overcome this problem, we propose an approach called Locality Preserving Feature Learning (LPFL), which incorporates feature selection into LPI. Specifically, we aim to find a subset of features, and learn a linear transformation to optimize the Locality Preserving Criterion based on these features. The resulting optimization problem is a mixed integer programming problem, which we relax into a constrained Frobenius norm minimization problem, and solve using a variation of Alternating Direction Method (ADM). ADM, which iteratively updates the linear transformation matrix, the residue matrix and the Lagrangian multiplier, is theoretically guaranteed to converge at the rate O(1/t). Experiments on benchmark document datasets show that our proposed method outperforms LPI, as well as other state-of-the-art document analysis approaches.} }
Endnote
%0 Conference Paper %T Locality Preserving Feature Learning %A Quanquan Gu %A Marina Danilevsky %A Zhenhui Li %A Jiawei Han %B Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2012 %E Neil D. Lawrence %E Mark Girolami %F pmlr-v22-gu12 %I PMLR %J Proceedings of Machine Learning Research %P 477--485 %U http://proceedings.mlr.press %V 22 %W PMLR %X Locality Preserving Indexing (LPI) has been quite successful in tackling document analysis problems, such as clustering or classification. The approach relies on the Locality Preserving Criterion, which preserves the locality of the data points. However, LPI takes every word in a data corpus into account, even though many words may not be useful for document clustering. To overcome this problem, we propose an approach called Locality Preserving Feature Learning (LPFL), which incorporates feature selection into LPI. Specifically, we aim to find a subset of features, and learn a linear transformation to optimize the Locality Preserving Criterion based on these features. The resulting optimization problem is a mixed integer programming problem, which we relax into a constrained Frobenius norm minimization problem, and solve using a variation of Alternating Direction Method (ADM). ADM, which iteratively updates the linear transformation matrix, the residue matrix and the Lagrangian multiplier, is theoretically guaranteed to converge at the rate O(1/t). Experiments on benchmark document datasets show that our proposed method outperforms LPI, as well as other state-of-the-art document analysis approaches.
RIS
TY - CPAPER TI - Locality Preserving Feature Learning AU - Quanquan Gu AU - Marina Danilevsky AU - Zhenhui Li AU - Jiawei Han BT - Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics PY - 2012/03/21 DA - 2012/03/21 ED - Neil D. Lawrence ED - Mark Girolami ID - pmlr-v22-gu12 PB - PMLR SP - 477 DP - PMLR EP - 485 L1 - http://proceedings.mlr.press/v22/gu12/gu12.pdf UR - http://proceedings.mlr.press/v22/gu12.html AB - Locality Preserving Indexing (LPI) has been quite successful in tackling document analysis problems, such as clustering or classification. The approach relies on the Locality Preserving Criterion, which preserves the locality of the data points. However, LPI takes every word in a data corpus into account, even though many words may not be useful for document clustering. To overcome this problem, we propose an approach called Locality Preserving Feature Learning (LPFL), which incorporates feature selection into LPI. Specifically, we aim to find a subset of features, and learn a linear transformation to optimize the Locality Preserving Criterion based on these features. The resulting optimization problem is a mixed integer programming problem, which we relax into a constrained Frobenius norm minimization problem, and solve using a variation of Alternating Direction Method (ADM). ADM, which iteratively updates the linear transformation matrix, the residue matrix and the Lagrangian multiplier, is theoretically guaranteed to converge at the rate O(1/t). Experiments on benchmark document datasets show that our proposed method outperforms LPI, as well as other state-of-the-art document analysis approaches. ER -
APA
Gu, Q., Danilevsky, M., Li, Z. & Han, J.. (2012). Locality Preserving Feature Learning. Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, in PMLR 22:477-485

Related Material