A family of one-class classification methods is extended by the determinant maximization novelty detection (DMND) model based on the D-optimum experimental design approach for the ellipsoid estimation. Similar to the one-class classification methods based on the support vector machine or the so-called support vector data description (SVDD) approach, DMND is a method that fits a geometrical object around the training data. However, in contrast to SVDD, DMND finds the hyperellipsoid of the smallest volume covering the target objects that can contain outliers by maximizing the determinant of an information matrix. Simulation results are presented for the case when training data are contaminated by compactly located outliers.