Graph-Embedded Subspace-Learning for stress level detection

A novel subspace learning framework for obtaining a mapping along with data description in low-dimensional feature space. The framework presents the problem of subspace learning for one-class classification in the form of graph embedding. It includes the previously proposed subspace one-class techniques as its special cases and provides further insight into what these techniques optimize. The framework reveals a spectral solution and a spectral regression-based solution as alternatives to the previously used gradient-based technique.

The framework places subspace learning for SVDD in the general graph embedding framework with a fixed data-dependent SVDD graph , which resembles PCA on the support vectors and outliers, and an additional constraint graph , which allows to incorporate other meaningful data relationships to the subspace learning step. In the earlier works, the overall objective function has been minimized via gradient-descent. However, the new framework hints that it can also make sense to reverse the objective and maximize instead of minimizing. The system for facial expression recognition has the following two primary steps. • The face of the subject is detected in the image, and relevant features are extracted. • A classifier is used to predict the emotions based on extracted features.

Useable in variuos programming languages, applications and platforms, e.g., MATLAB.

Hospitals and Healthcare providers.

Open access under the CC BY license (http://creativecommons.org/licenses/by/4.0/)