Data stream classification: a review

Wankhade, Kapil; Dongre, Snehlata; Jondhale, Kalpana C.

doi:10.1007/s42044-020-00061-3

Cited by 18 publications

(7 citation statements)

References 123 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is noticed that the average accuracy of all algorithms varies between 0.89 and 0.87 success rate. Because these results are very close, a statistical assessment of the results is Dataset ARF HAT AXGB AFXGB Agrawal-A 0.9659 (4) 0.9808 (3) 0.9868 (2) 0.9872 (1) Agrawal-G 0.955 (4) 0.9649 (3) 0.9724 (2) 0.9786 (1) Hyperplanes 0.8485 (4) 0.8609 (1) 0.857 (2) 0.852 (3) SEA-A 0.9956 (1) 0.9879 (2) 0.986 (4) 0.9871 (3) SEA-G 0.9923 (1) 0.986 (2) 0.9857 (4) 0.9859 (3) Airlines 0.6722 (1) 0.64 (2) 0.6327 (3) 0.6312 (4) Electricity Market 0.863 (1) 0.8133 (2) 0.7273 (3) needed [18]. To find if there is a performance difference between the algorithms, we start applying the Friedmann test.…”

Section: Accuracy Analysismentioning

confidence: 99%

“…Adding this to the broad use of the Internet of Things and smart devices, we witness the shaping of the so-called Big Data, a mass of information characterized by the volume, variety, and speed with which the data is presented [1]. In this scenario, data is increasingly being used for machine learning, a technique that is applied in several real-world classification tasks [2,3].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Fast XGBoost for Binary Classification

Baldo

Grando

Weege

et al. 2022

Anais Do XXXVII Simpósio Brasileiro De Banco De Dados (SBBD 2022)

View full text Add to dashboard Cite

Modern machine learning algorithms must be able to fast consume data streams, maintaining accurate results, even with the presence of concept drift. This work proposes AFXGB, an Adaptive Fast binary classification algorithm using XGBoost, focusing on the fast induction of labeled data streams. AFXGB uses an alternate model training strategy to achieve lean models adapted to concept drift. We compared AFXGB with other data stream classifiers using synthetic and real datasets. The results showed that AFXGB is four times faster than ARF and 22 times faster than AXGB, maintaining the same accuracy and with the fastest recovery from concept drifts, thus preserving long-term accuracy.

show abstract

Section: Accuracy Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Adaptive Fast XGBoost for Binary Classification

Baldo

Grando

Weege

et al. 2022

Anais Do XXXVII Simpósio Brasileiro De Banco De Dados (SBBD 2022)

View full text Add to dashboard Cite

show abstract

“…For example, AdaDeepStream [5], class-based ensemble approach for class evolution (CBCE) [25] and CLAss-based Micro classifier ensemble (CLAM) [1]. For more details about concept evolution, see [9,17,29].…”

Section: Introductionmentioning

confidence: 99%

Online learning for streaming data classification in nonstationary environments

Gai,

Meng,

Wang

2024

Statistical Analysis

View full text Add to dashboard Cite

In this article, we implement the classification of nonstationary streaming data. Due to the inability to obtain full data in the context of streaming data, we adopt a strategy based on clustering structure for data classification. Specifically, this strategy involves dynamically maintaining clustering structures to update the model, thereby updating the objective function for classification. Simultaneously, incoming samples are monitored in real‐time to identify the emergence of new classes or the presence of outliers. Moreover, this strategy can also deal with the concept drift problem, where the distribution of data changes with the inflow of data. Regarding the handling of novel instances, we introduce a buffer analysis mechanism to delay their processing, which in turn improves the prediction performance of the model. In the process of model updating, we also introduce a novel renewable strategy for the covariance matrix. Numerical simulations and experiments on datasets show that our method has significant advantages.

show abstract

“…Online classification is thus viewed as a viable technique for classifying electricity consumption patterns in smart grid systems because user behavior may change over time in an unanticipated way. Numerous techniques have been put forth for the classification of data streams [11]- [13].…”

Section: Introductionmentioning

confidence: 99%

Evaluation of Online Machine Learning Algorithms for Electricity Theft Detection in Smart Grids

Alkhresheh¹,

Al-Tarawneh²,

Alnawayseh³

2022

IJACSA

View full text Add to dashboard Cite

Electricity theft-induced power loss is a pressing issue in both traditional and smart grid environments. In smart grids, smart meters can be used to track power consumption behaviour and detect any suspicious activity. However, smart meter readings can be compromised by deploying intrusion tactics or launching cyber attacks. In this regard, machine learning models can be used to assess the daily consumption patterns of customers and detect potential electricity theft incidents. Whilst existing research efforts have extensively focused on batch learning algorithms, this paper investigates the use of online machine learning algorithms for electricity theft detection in smart grid environments, based on a recently proposed dataset. Several algorithms including Naive Bayes, K-nearest Neighbours, K-nearest Neighbours with self-adjusting memory, Hoeffding Tree, Extremely Fast Decision Tree, Adaptive Random Forest and Leveraging Bagging are considered. These algorithms are evaluated using an online machine learning platform considering both binary and multi-class theft detection scenarios. Evaluation metrics include prediction accuracy, precision, recall, F-1 score and kappa statistic. Evaluation results demonstrate the ability of the Leveraging Bagging algorithm with an Adaptive Random Forest base classifier to surpass all other algorithms in terms of all the considered metrics, for both binary and multi-class theft detection. Hence, it can be considered as a viable option for electricity theft detection in smart grid environments.

show abstract

Data stream classification: a review

Cited by 18 publications

References 123 publications

Adaptive Fast XGBoost for Binary Classification

Adaptive Fast XGBoost for Binary Classification

Online learning for streaming data classification in nonstationary environments

Evaluation of Online Machine Learning Algorithms for Electricity Theft Detection in Smart Grids

Contact Info

Product

Resources

About