A class-driven approach to dimension embedding

作者：

Highlights：

• The proposed algorithm called CDE carries out an exact embedding for test points.

• CDE does not suffer from outliers due to utilizing class information well.

• A class-aware embedding provides more explanatory information about the classes.

• A new distance metric based on information gain is proposed.

• The embedded data delivers a higher accuracy than originals in many data sets.

摘要

•The proposed algorithm called CDE carries out an exact embedding for test points.•CDE does not suffer from outliers due to utilizing class information well.•A class-aware embedding provides more explanatory information about the classes.•A new distance metric based on information gain is proposed.•The embedded data delivers a higher accuracy than originals in many data sets.

论文关键词：Machine learning,Dimension embedding,Cumulative distribution function,Brownian motion,Concave-convex functions,Distance metrics

论文评审过程：Received 6 April 2021, Revised 21 January 2022, Accepted 4 February 2022, Available online 11 February 2022, Version of Record 16 February 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116650