PROJECT TITLE :

A Low-Rank Approximation-Based Transductive Support Tensor Machine for Semisupervised Classification

ABSTRACT:

Within the fields of machine learning, pattern recognition, image processing, and laptop vision, the data are sometimes represented by the tensors. For the semisupervised tensor classification, the prevailing transductive support tensor machine (TSTM) wants to resort to iterative technique, that is terribly time-consuming. In order to beat this shortcoming, during this paper, we tend to extend the concave–convex procedure-based transductive support vector machine (CCCP-TSVM) to the tensor patterns and propose a coffee-rank approximation-based mostly TSTM, in which the tensor rank-one decomposition is used to compute the inner product of the tensors. Theoretically, concave–convex procedure-primarily based TSTM (CCCP-TSTM) is an extension of the linear CCCP-TSVM to tensor patterns. When the input patterns are vectors, CCCP-TSTM degenerates into the linear CCCP-TSVM. A group of experiments is conducted on twenty three semisupervised classification tasks, that are generated from seven second-order face knowledge sets, 3 third-order gait knowledge sets, and two third-order image data sets, to illustrate the performance of the CCCP-TSTM. The results show that compared with CCCP-TSVM and TSTM, CCCP-TSTM provides important performance gain in terms of test accuracy and training speed.