Rapid advances in technology have made classification with high dimensional features and ubiquitous problem in modern scientific studies and applications. There are three fundamental goals in the pursuit of a good high-dimensional classifier: accuracy, interpretability, and scalability. In the past 15 years, a host of competitive high-dimensional classifiers have been developed based on sparse regularization techniques. In this article, we give a selective overview of these classification methods. This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Knowledge Discovery.
|Original language||English (US)|
|Journal||Wiley Interdisciplinary Reviews: Computational Statistics|
|State||Published - Jan 1 2019|
Bibliographical noteFunding Information:
This work is supported by NSF grant DMS-1505111. The author thanks three referees for their helpful comments and suggestions.
© 2018 Wiley Periodicals, Inc.
Copyright 2018 Elsevier B.V., All rights reserved.
- feature selection