Exploring Concept Representations for Concept Drift Detection.

Oliver Becher; Laura Hollink; Desmond Elliott

Exploring Concept Representations for Concept Drift Detection.

Oliver Becher, Laura Hollink, Desmond Elliott

3 Citations (Scopus)

Abstract

We present an approach to estimating concept drift in online news. Our method is to construct temporal concept vectors from topicannotated news articles, and to correlate the distance between the temporal concept vectors with edits to the Wikipedia entries of the concepts. We find improvements in the correlation when we split the news articles based on the amount of articles mentioning a concept, instead of calendar-based units of time.

Original language	Undefined/Unknown
Title of host publication	SEMANTICS Workshops
Publication date	2017
Publication status	Published - 2017

Cite this

@inproceedings{0a17472a638e433696fec95a9a39db5b,

title = "Exploring Concept Representations for Concept Drift Detection.",

abstract = "We present an approach to estimating concept drift in online news. Our method is to construct temporal concept vectors from topicannotated news articles, and to correlate the distance between the temporal concept vectors with edits to the Wikipedia entries of the concepts. We find improvements in the correlation when we split the news articles based on the amount of articles mentioning a concept, instead of calendar-based units of time.",

author = "Oliver Becher and Laura Hollink and Desmond Elliott",

year = "2017",

language = "Udefineret/Ukendt",

booktitle = "SEMANTICS Workshops",

}

TY - GEN

T1 - Exploring Concept Representations for Concept Drift Detection.

AU - Becher, Oliver

AU - Hollink, Laura

AU - Elliott, Desmond

PY - 2017

Y1 - 2017

N2 - We present an approach to estimating concept drift in online news. Our method is to construct temporal concept vectors from topicannotated news articles, and to correlate the distance between the temporal concept vectors with edits to the Wikipedia entries of the concepts. We find improvements in the correlation when we split the news articles based on the amount of articles mentioning a concept, instead of calendar-based units of time.

AB - We present an approach to estimating concept drift in online news. Our method is to construct temporal concept vectors from topicannotated news articles, and to correlate the distance between the temporal concept vectors with edits to the Wikipedia entries of the concepts. We find improvements in the correlation when we split the news articles based on the amount of articles mentioning a concept, instead of calendar-based units of time.

M3 - Konferencebidrag i proceedings

BT - SEMANTICS Workshops

ER -