Detecting Different Types of Concept Drifts with Ensemble Framework

Hu, Hanqing; Kantardzic, Mehmed; Lyu, Lingyu

doi:10.1109/icmla.2018.00058

Cited by 9 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An approach, named ensemble of drift detection (EFDD) proposed by Hu et al (2018), aims to detect all types of concept drift in an unlabeled data stream by combining the advantage of several distribution-based concept drift detection methods. Concept drift detection strategies from MD3 (Sethi & Kantardzic, 2017), clustering (Kantardzic et al, 2010) (cluster densitybased detection) and GC3 (Sethi et al, 2016) (grid density-based detection) are integrated into an ensemble.…”

Section: Detecting Different Types Of Concept Drift Using Ensemble mentioning

confidence: 99%

“…While unlabeled data can be used for concept drift detection, eliminating the cost of human intervention (Cabanes & Bennani, 2012;dos Reis, Flach, Matwin, & Batista, 2016), this however, often results in lowered detection performance, since concept drifts take on various forms and are often unpredictable (Hu, Kantardzic, & Lyu, 2018). Two examples of changes in unlabeled data stream are shown in Figure 3.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

No Free Lunch Theorem for concept drift detection in streaming data classification: A review

Kantardzic

Sethi

2019

WIREs Data Min & Knowl

Self Cite

View full text Add to dashboard Cite

Many real‐world data mining applications have to deal with unlabeled streaming data. They are unlabeled because the sheer volume of the stream makes it impractical to label a significant portion of the data. The data streams can evolve over time and these changes are called concept drifts. Concept drifts have different characteristics, which can be used to categorize them into different types. A trade‐off between performance and cost exists among many concept drift detection approaches. On the one hand, high accuracy detection approach usually requires labeled data, possibly involving high cost for labeling. On the other hand, a variety of methods have been devoted to the topic of concept drift detection with unlabeled data, but these approaches often are most suited for only a subset of the concept drift types. The objective of this survey is to present these methods, categorize them and give recommendations of usage based on their behaviors under different types of concept drift. This article is categorized under: Fundamental Concepts of Data and Knowledge > Data Concepts Fundamental Concepts of Data and Knowledge > Key Design Issues in Data Mining Explainable AI > Classification

show abstract

Section: Detecting Different Types Of Concept Drift Using Ensemble mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

No Free Lunch Theorem for concept drift detection in streaming data classification: A review

Kantardzic

Sethi

2019

WIREs Data Min & Knowl

Self Cite

View full text Add to dashboard Cite

show abstract

“…It can detect abrupt drift and gradual drift at the same time and adopts a selective ensemble strategy to find concept drifts that cannot be detected by a single method based on change indicators. The drift detector ensemble framework (EFDD) proposed by Hu et al 89 uses type‐based voting. For a single algorithm covering the same type of concept drift, the majority vote will determine the detection result.…”

Section: Active Handling Methodsmentioning

confidence: 99%

A survey of active and passive concept drift handling methods

Han

Chen

et al. 2022

Computational Intelligence

View full text Add to dashboard Cite

At present, concept drift in the nonstationary data stream is showing trends with different speeds and different degrees of severity, which has brought great challenges to many fields like data mining and machine learning. In the past two decades, a lot of methods dedicated to handling concept drift in the nonstationary data stream have emerged. A novel perspective is proposed to classify these methods, and the current concept drift handling methods are comprehensively explained from the active handling methods and the passive handling methods. In particular, active handling methods are analyzed from the perspective of handling one specific type of concept drift and handling multiple types of concept drift, and passive handling methods are analyzed from the perspective of single learner and ensemble learning. Many concept drift handling methods in this survey are analyzed and summarized in terms of the comparing algorithms, learning model, applicable drift type, advantages, and disadvantages of the algorithms. Finally, further research directions are given, including the active and passive mixing methods, class imbalance, the existence of novel class in the data stream, and the noise in the data stream.

show abstract