Experiments with two Approaches for Tracking Drifting Concepts

Koychev, Ivan. "Experiments with two Approaches for Tracking Drifting Concepts." Serdica Journal of Computing 1.1 (2007): 27-44. <http://eudml.org/doc/11410>.

@article{Koychev2007,
abstract = {This paper addresses the task of learning classifiers from streams of labelled data. In this case we can face the problem that the underlying concepts can change over time. The paper studies two mechanisms developed for dealing with changing concepts. Both are based on the time window idea. The first one forgets gradually, by assigning to the examples weight that gradually decreases over time. The second one uses a statistical test to detect changes in concept and then optimizes the size of the time window, aiming to maximise the classification accuracy on the new examples. Both methods are general in nature and can be used with any learning algorithm. The objectives of the conducted experiments were to compare the mechanisms and explore whether they can be combined to achieve a synergetic e ect. Results from experiments with three basic learning algorithms (kNN, ID3 and NBC) using four datasets are reported and discussed.},
author = {Koychev, Ivan},
journal = {Serdica Journal of Computing},
keywords = {Machine Learning; Concept Drift; Forgetting Models; learning classifiers},
language = {eng},
number = {1},
pages = {27-44},
publisher = {Institute of Mathematics and Informatics Bulgarian Academy of Sciences},
title = {Experiments with two Approaches for Tracking Drifting Concepts},
url = {http://eudml.org/doc/11410},
volume = {1},
year = {2007},
}

TY - JOUR
AU - Koychev, Ivan
TI - Experiments with two Approaches for Tracking Drifting Concepts
JO - Serdica Journal of Computing
PY - 2007
PB - Institute of Mathematics and Informatics Bulgarian Academy of Sciences
VL - 1
IS - 1
SP - 27
EP - 44
AB - This paper addresses the task of learning classifiers from streams of labelled data. In this case we can face the problem that the underlying concepts can change over time. The paper studies two mechanisms developed for dealing with changing concepts. Both are based on the time window idea. The first one forgets gradually, by assigning to the examples weight that gradually decreases over time. The second one uses a statistical test to detect changes in concept and then optimizes the size of the time window, aiming to maximise the classification accuracy on the new examples. Both methods are general in nature and can be used with any learning algorithm. The objectives of the conducted experiments were to compare the mechanisms and explore whether they can be combined to achieve a synergetic e ect. Results from experiments with three basic learning algorithms (kNN, ID3 and NBC) using four datasets are reported and discussed.
LA - eng
KW - Machine Learning; Concept Drift; Forgetting Models; learning classifiers
UR - http://eudml.org/doc/11410
ER -