Announcing the Candidacy Exam of Emil Agbemade for the PhD in Big Data Analytics
Significant attention has been drawn to support vector data description (SVDD) as a result of its exceptional performance in one-class classification and novelty detection tasks. Nevertheless, during the modelling process, all slack variables are assigned the same weight. This can lead to a decline in learning performance if the training data contains erroneous observations or outliers. In this study, an extended SVDD model (RSVDD) is introduced to strengthen the resistance of the traditional SVDD to anomalies. This is achieved by redefining the initial optimization problem of SVDD using a hinge loss function that has been rescaled. As this loss function can increase the significance of samples that are more likely to represent the target class while decreasing the impact of samples that are more likely to represent anomalies, it can be considered a weighted SVDD. To efficiently address the optimization challenge associated with the proposed model, the half-quadratic optimization method was utilized to generate a dynamic optimization algorithm. Experimental findings on a synthetic data set are presented to illustrate the new proposed method's performance superiority over SVDD.
Committee in Charge:
Dr. Edgard Maboudou, Chair
Dr. Alexander Mantzaris
Dr. Larry Tang
Dr. Helen Huang
Read More