MWFP-outlier: Maximal weighted frequent-pattern-based approach for detecting outliers from uncertain weighted data streams

Cai, Saihua; Li, Li; Chen, Jinfu; Zhao, Kaiyi; Yuan, Gang; Sosu, Rexford Nii Ayitey; Huang, Liusheng

doi:10.1016/j.ins.2022.01.028

Cited by 9 publications

(9 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6(a). In contrast, other methods have achieved lower rates, which are 71.87%, 70.63%, 73.22%, 88.03%, 90.16%, 93.81%, and 84.32% of the recall measure for the methods UKOF [14], MWFP-Outlier [15], LiCS [21], iLDCBOF [22], Method in [26], Method in [27], and ASEC-OD [28], respectively. One thing is notable here is that the MWFP-Outlier and LiCS methods have the lowest performance rates on all datasets in most cases.…”

Section: E Results and Discussionmentioning

confidence: 95%

“…The proposed framework is evaluated by comparing results with its standalone base learners and other some predictive machine learning techniques such as Logistic Regression (LR), SVM, GaussianNB, K-NN, Artificial Neural Network (ANN), Gradient Boosting, and Local Outlier Factor (LOF). In addition, ESOD is compared with many state-ofthe-art methods found in [14], [15], [21], [22], [26], [27], and [28]. We individually show the performance of the comparisons on the aforementioned 11 datasets using the accuracy, precision, recall, and F1-score metrics.…”

Section: E Results and Discussionmentioning

confidence: 99%

“…The next set of experiments is performed to compare the proposed framework with some state-of-the-art methods for streaming outlier detection presented in [14], [15], [21], [22], [26], [27] and [28] and the results are shown in Fig 6 . In more detail, from Fig. 6(a)-(k), the performance of the proposed model (ESOD) is better than its competitive methods on all benchmark datasets, where it achieves higher rates than others.…”

Section: E Results and Discussionmentioning

confidence: 99%

“…Aside from improving execution performance, we are also interested in implementing various time-based techniques that may lead to improved detection and accuracy rates. UKOF [14] MWFP-Outlier [15] LiCS [21] iLDCBOF [22] Method in [26] Method in [27] ASEC-OD [28] ESOD…”

Section: Discussionmentioning

confidence: 99%

“…Although this method had a low computational cost, it assumed a stable distribution of data, which is inappropriate in a data stream setting. In recent www.ijacsa.thesai.org research [15], a maximal weighted frequent-pattern-based outlier detection approach (MWFP-Outlier) was proposed to detect potential outliers from uncertain data streams. This approach had two phases, namely pattern mining phase and an outlier detection phase.…”

Section: Related Workmentioning

confidence: 99%

See 4 more Smart Citations

An Effective Ensemble-based Framework for Outlier Detection in Evolving Data Streams

Hassan¹,

Barakat²,

Rezk³

2022

IJACSA

View full text Add to dashboard Cite

In the last few years, data streams have drawn lots of researchers' attention due to their various applications, such as healthcare monitoring systems, fraud and intrusion detection, the internet of things (IoT), and financial market applications. A data stream is an unbounded sequence of data continually generated over time and is prone to evolution. Outliers in streaming data are the elements that significantly deviate from the majority of elements and then have to be detected as they may be error values or events of interest. Detection of outliers is a challenging issue in streaming data and is one of the most crucial tasks in data stream mining. Existing outlier detection methods for static data are unsuitable for use in data stream settings due to the unique characteristics of streaming data such as unpredictability, uncertainty, high-dimensionality, and changes in data distribution. Thus, in this paper, a novel ensemble learning framework called Ensemble-based Streaming Outlier Detection (ESOD) is presented to perfectly detect outliers over streaming data using a sliding window technique that is updated in response to the incoming events from the data streaming environment to overcome the concept evolution nature of streaming data. The proposed framework has three phases, namely the training phase, testing/offline phase, and outlier detection/online phase. A detection weighted vote technique is used to determine the final decisions for potential outliers. In the extensive experimental study, which was conducted on 11 realworld benchmark datasets, the proposed framework was assessed using many accuracy metrics. The experiment results showed that the proposed framework beats many other state-ofthe-art methods.

show abstract