Feature Selection for Handling Concept Drift in the Data Stream Classification

Turkov, Pavel A.; Krasotkina, Olga; Mottl, Vadim; Sychugov, Alexey

doi:10.1007/978-3-319-41920-6_48

Cited by 6 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Subsequently, we propose the DLVSW-CDTD algorithms to effectively detect different types of CD during the data stream mining process. In the fourth section, extensive experiments are conducted using real and synthetic datasets obtained using the open-source data mining library SPMF [25]. The results confirm the efficiency and feasibility of the algorithm.…”

Section: Introductionmentioning

confidence: 59%

“…We use a random transactional data generator provided in an open source data mining library SPMF [25] to generate a comprehensive transactional data set. The relevant information of the data set is introduced as follows.…”

Section: Open Source Data Mining Library Spmfmentioning

confidence: 99%

“…Riverola et al demonstrated an improvement to a successful e-mail filtering model to track spam domain CD [23], while Gulla et al discussed a new approach to detect semantic drift using concept signatures [24]. A new Bayesian framework [25] in data stream pattern recognition was presented for feature selection. Ruano-Ordás et al presented a detailed study of CD in the e-mail domain considering types of CD and message classes (spam and ham) [26].…”

Section: Introductionmentioning

confidence: 99%

“…A random transactional data generator provided in the open source data mining library SPMF [25] was used to generate a comprehensive transactional data set, as follows.…”

Section: Introduction Of the Experimental Environment And Data Set Ex...mentioning

confidence: 99%

See 3 more Smart Citations

Multi-type concept drift detection under a dual-layer variable sliding window in frequent pattern mining with cloud computing

Chen,

Yang,

Gao

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

The detection of different types of concept drift has wide applications in the fields of cloud computing and security information detection. Concept drift detection can indeed assist in promptly identifying instances where model performance deteriorates or when there are changes in data distribution. This paper focuses on the problem of concept drift detection in order to conduct frequent pattern mining. To address the limitation of fixed sliding windows in adapting to evolving data streams, we propose a variable sliding window frequent pattern mining algorithm, which dynamically adjusts the window size to adapt to new concept drifts and detect them in a timely manner. Furthermore, considering the challenge of existing concept drift detection algorithms that struggle to adapt to different types of drifting data simultaneously, we introduce an additional dual-layer embedded variable sliding window. This approach helps differentiate types of concept drift and incorporates a decay model for drift adaptation. The proposed algorithm can effectively detect different types of concept drift in data streams, perform targeted drift adaptation, and exhibit efficiency in terms of time complexity and memory consumption. Additionally, the algorithm maintains stable performance, avoiding abrupt changes due to window size variations and ensuring overall robustness.

show abstract

Section: Introductionmentioning

confidence: 59%

Section: Open Source Data Mining Library Spmfmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…A random transactional data generator provided in the open source data mining library SPMF [25] was used to generate a comprehensive transactional data set, as follows.…”

Section: Introduction Of the Experimental Environment And Data Set Ex...mentioning

confidence: 99%

See 2 more Smart Citations

Multi-type concept drift detection under a dual-layer variable sliding window in frequent pattern mining with cloud computing

Chen,

Yang,

Gao

et al. 2024

J Cloud Comp

View full text Add to dashboard Cite

show abstract

“…It is integrated with the Waikato Environment for Knowledge Analysis (WEKA) and provides a group of native tools to evaluate different algorithms in streaming environments. We selected this framework because of its widespread use in the literature - [Bifet et al 2010], [Turkov et al 2016] and [Ramírez-Gallego et al 2017] -, its userfriendly interface, for being open-source, and, at last, for its built-in tools to evaluate algorithms and procedures of machine learning when applied to data streams, such as classifiers, feature selectors, predictors, and regressors. We choose the following versions: MOA 2016.04 and WEKA 3.8.…”

Section: Methodsmentioning

confidence: 99%

Performance Evaluation of Feature Selection Algorithms Applied to Online Learning in Concept Drift Environments

Moraes¹,

Gradvohl²

2018

Anais Do XV Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2018)

View full text Add to dashboard Cite

Data streams are transmitted at high speeds with huge volume and may contain critical information need processing in real-time. Hence, to reduce computational cost and time, the system may apply a feature selection algorithm. However, this is not a trivial task due to the concept drift. In this work, we show that two feature selection algorithms, Information Gain and Online Feature Selection, present lower performance when compared to classification tasks without feature selection. Both algorithms presented more relevant results in one distinct scenario each, showing final accuracies up to 14% higher. The experiments using both real and artificial datasets present a potential for using these methods due to their better adaptability in some concept drift situations.

show abstract

Deep Incremental Learning for Big Data Stream Analytics

Alex

Nayahi²

2019

Lecture Notes on Data Engineering and Communications Technologies

View full text Add to dashboard Cite

Feature Selection for Handling Concept Drift in the Data Stream Classification

Cited by 6 publications

References 17 publications

Multi-type concept drift detection under a dual-layer variable sliding window in frequent pattern mining with cloud computing

Multi-type concept drift detection under a dual-layer variable sliding window in frequent pattern mining with cloud computing

Performance Evaluation of Feature Selection Algorithms Applied to Online Learning in Concept Drift Environments

Deep Incremental Learning for Big Data Stream Analytics

Contact Info

Product

Resources

About