Depth-based classification of directional data

• Published in: Expert systems with applications Vol. 169; p. 114433
• Main Authors: Pandolfo, Giuseppe, D’Ambrosio, Antonio
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.05.2021; Elsevier BV
• Subjects: Angular data depth; Classification; Directional distance; Misclassification rate; Spherical random variables; Misclassification rate; Spherical random variables; Angular data depth; Directional distance
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2020.114433

A non-parametric procedure based on the concept angular depth function is developed for dealing with classification problems of objects in directional statistics. Several notions of depth for directional data are adopted: the angular simplicial, the angular Tukey’s, the arc distance, the cosine distance and the chord distance depths. The proposed method is flexible and can be applied even in high-dimensional cases when a suitable notion of depth is adopted. Performances are investigated and compared by applying methods to different distributional settings through simulated and real data sets. •A non-parametric spherical-distance-based classifier is proposed.•It is an alternative to the existing depth-based algorithms.•Performances are investigated and compared through simulated and real data sets.