Analysis of Drifting Features

12/01/2020
by   Fabian Hinder, et al.
0

The notion of concept drift refers to the phenomenon that the distribution, which is underlying the observed data, changes over time. We are interested in an identification of those features, that are most relevant for the observed drift. We distinguish between drift inducing features, for which the observed feature drift cannot be explained by any other feature, and faithfully drifting features, which correlate with the present drift of other features. This notion gives rise to minimal subsets of the feature space, which are able to characterize the observed drift as a whole. We relate this problem to the problems of feature selection and feature relevance learning, which allows us to derive a detection algorithm. We demonstrate its usefulness on different benchmarks.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 9

12/04/2019

A probability theoretic approach to drifting data in continuous time domains

The notion of drift refers to the phenomenon that the distribution, whic...
06/23/2020

Counterfactual Explanations of Concept Drift

The notion of concept drift refers to the phenomenon that the distributi...
08/27/2020

Reducing Drift in Structure from Motion using Extended Features

Low-frequency long-range errors (drift) are an endemic problem in 3D str...
12/25/2012

Exponentially Weighted Moving Average Charts for Detecting Concept Drift

Classifying streaming data requires the development of methods which are...
10/22/2010

Learning under Concept Drift: an Overview

Concept drift refers to a non stationary learning problem over time. The...
10/08/2021

How to Do Things without Words: Modeling Semantic Drift of Emoji

Emoji have become a significant part of our informal textual communicati...
09/04/2019

Multi-DoF Time Domain Passivity Approach based Drift Compensation for Telemanipulation

When, in addition to stability, position synchronization is also desired...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.