Twin Kernel Embedding with Back Constraints

Yi Guo, Paul Kwan, Junbin Gao

Research output: Book chapter/Published conference paperConference paperpeer-review

3 Citations (Scopus)
4 Downloads (Pure)


Twin kernel embedding (TKE) is a novel approach for visualization of non-vectorial objects. It preserves the similarity structure in high-dimensional or structured input data and reproduces it in a low dimensional latent space by matching the similarity relations represented by two kernel gram matrices, one kernel for the input data and the other for embedded data. However, there is no explicit mapping from the input data to their corresponding low dimensional embeddings. We obtain this mapping by including the back constraints on the data in TKE in this paper. This procedure still emphasizes the locality preserving. Further, the smooth mapping also solves the problem of so-called out-of-sample problem which is absent in the original TKE. Experimental evaluation on different real world data sets verifies the usefulness of this method.
Original languageEnglish
Title of host publicationIEEE International Conference on Data Mining Workshops (ICDMW)
EditorsYong Shi Shi, Chris Clifton
Place of PublicationWashington DC, USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)9780769530192
Publication statusPublished - 2007
Event7th International Conference - Omaha, USA, New Zealand
Duration: 28 Oct 200731 Oct 2007


Conference7th International Conference
CountryNew Zealand

Fingerprint Dive into the research topics of 'Twin Kernel Embedding with Back Constraints'. Together they form a unique fingerprint.

Cite this