Abstract | ||
---|---|---|
Concept drift involving noise is an important research in the field of data mining. Many concept drift detection models are proposed to promote the research of traditional concept drift detection. In this paper, we propose an anti-noise concept drift processing algorithm based on entropy of information, named ACPJS. In ACPJS, the JS-divergence and Hoeffding Bounds are used to set double threshold for concept drift detection and subsequently a horizontal integrated model will be constructed for anti-noise concept drift processing. In the comparison experiments of multiple data sets, the presented algorithm has shown good performance in concept drift detection, anti-noise performance and classification accuracy. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1007/978-3-030-22796-8_47 | ADVANCES IN NEURAL NETWORKS - ISNN 2019, PT I |
Keywords | Field | DocType |
Concept drift, JS-divergence, Horizontal integrated model | Multiple data,Divergence,Computer science,Algorithm,Concept drift,Double threshold | Conference |
Volume | ISSN | Citations |
11554 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xin Song | 1 | 15 | 15.82 |
Shizhen Qin | 2 | 0 | 0.34 |
Shaokai Niu | 3 | 0 | 0.34 |
Yan Wang | 4 | 2 | 1.07 |