PROJECT TITLE :

Unsupervised domain adaptation using eigenanalysis in kernel space for categorisation tasks

ABSTRACT:

This study describes a new technique of unsupervised domain adaptation primarily based on eigenanalysis in kernel house, for the aim of categorisation tasks. The authors propose a change of data in supply domain, such that the eigenvectors and eigenvalues of the reworked supply domain become just like that of the target domain. They extend this concept to the reproducing kernel Hilbert space, that allows to accommodate non-linear transformation of source domain. They conjointly propose a measure to get the appropriate number of eigenvectors needed for transformation. Results on object, video and text categorisations tasks using real-world datasets show that the proposed method produces higher results in comparison with a few recent state-of-art methods of domain adaptation.