In processing stream data, time is one of the most significant facts not only because the size of data is dramatically increased but because the characteristics of data is varying over time. To learn stream data evolving over time effectively, it is required to detect the drift of concept. We present a window adaptation function on domain value (WAV) to determine the size of windowed batch for learning algorithms of stream data and a method to detect the change of data characteristics with a criterion function utilizing correlation. When applying our adaptation function to a clustering task on a multi-stream data model, the result of learning synopsis of windowed batch determined by it shows its effectiveness. Our criterion function with correlation information of value distribution over time can be the reasonable threshold to detect the change between windowed batches.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Jeonghoon LEE, Yoon-Joon LEE, "Concept Drift Detection for Evolving Stream Data" in IEICE TRANSACTIONS on Information,
vol. E94-D, no. 11, pp. 2288-2292, November 2011, doi: 10.1587/transinf.E94.D.2288.
Abstract: In processing stream data, time is one of the most significant facts not only because the size of data is dramatically increased but because the characteristics of data is varying over time. To learn stream data evolving over time effectively, it is required to detect the drift of concept. We present a window adaptation function on domain value (WAV) to determine the size of windowed batch for learning algorithms of stream data and a method to detect the change of data characteristics with a criterion function utilizing correlation. When applying our adaptation function to a clustering task on a multi-stream data model, the result of learning synopsis of windowed batch determined by it shows its effectiveness. Our criterion function with correlation information of value distribution over time can be the reasonable threshold to detect the change between windowed batches.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E94.D.2288/_p
Copy
@ARTICLE{e94-d_11_2288,
author={Jeonghoon LEE, Yoon-Joon LEE, },
journal={IEICE TRANSACTIONS on Information},
title={Concept Drift Detection for Evolving Stream Data},
year={2011},
volume={E94-D},
number={11},
pages={2288-2292},
abstract={In processing stream data, time is one of the most significant facts not only because the size of data is dramatically increased but because the characteristics of data is varying over time. To learn stream data evolving over time effectively, it is required to detect the drift of concept. We present a window adaptation function on domain value (WAV) to determine the size of windowed batch for learning algorithms of stream data and a method to detect the change of data characteristics with a criterion function utilizing correlation. When applying our adaptation function to a clustering task on a multi-stream data model, the result of learning synopsis of windowed batch determined by it shows its effectiveness. Our criterion function with correlation information of value distribution over time can be the reasonable threshold to detect the change between windowed batches.},
keywords={},
doi={10.1587/transinf.E94.D.2288},
ISSN={1745-1361},
month={November},}
Copy
TY - JOUR
TI - Concept Drift Detection for Evolving Stream Data
T2 - IEICE TRANSACTIONS on Information
SP - 2288
EP - 2292
AU - Jeonghoon LEE
AU - Yoon-Joon LEE
PY - 2011
DO - 10.1587/transinf.E94.D.2288
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E94-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2011
AB - In processing stream data, time is one of the most significant facts not only because the size of data is dramatically increased but because the characteristics of data is varying over time. To learn stream data evolving over time effectively, it is required to detect the drift of concept. We present a window adaptation function on domain value (WAV) to determine the size of windowed batch for learning algorithms of stream data and a method to detect the change of data characteristics with a criterion function utilizing correlation. When applying our adaptation function to a clustering task on a multi-stream data model, the result of learning synopsis of windowed batch determined by it shows its effectiveness. Our criterion function with correlation information of value distribution over time can be the reasonable threshold to detect the change between windowed batches.
ER -