Drift Detection for Multi-label Data Streams Based on Label Grouping and Entropy

Shi, Zhongwei; Wen, Yonggang; Feng, Chao; Zhang, Hai

doi:10.1109/icdmw.2014.92

Cited by 17 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, a batch-based incremental threshold interference technique was put forward to further tackle with the class imbalance issue by instance sharing. And Shi et al [116] adopted the EM method and Apriori algorithm to organize the class labels into different subsets where the labels are grouped based on the dependencies among them. And then these label subsets are considered as new atomic single class labels to be used to mark each arrived instance.…”

Section: ) Pt Based Multi-label Data Stream Classification Methodsmentioning

confidence: 99%

“…On the other hand, to tackle with concept drift issue, a label combination based method was presented in [116], which used the entropy of multi-label to measure the distribution of multi-label streaming data and detect concept drift. Some authors use dynamic ensemble [100], [108], [109] to adapt to concept drifts, which updates base classifiers in an ensemble or adjusts the concerning parameters of the base classifiers to track the latest concept.…”

Section: ) Ensemble Multi-label Data Stream Classification Methodsmentioning

confidence: 99%

“…And in [117], based on a basic generator, there is another multi-label stream data generation framework. And a common method is introduced in [105], [116], [120], which models concept drifts by creating a new multi-label data stream with shuffling labels. Contrary to the aforementioned methods, authors in [35] consider two categories of label dependencies from the angle of probability, namely conditional dependence and unconditional dependence.…”

Section: Synthetic Datasetsmentioning

confidence: 99%

“…Moreover, it is also able to simulate how this dependence and relationship evolve with the change of time. As far as we know, the most often used approach, in [115], [116], [121], is the MOA (Massive Online Analysis) framework, on which we could synthesize the needing dataset with different concept drifts and label dependencies.…”

Section: Synthetic Datasetsmentioning

confidence: 99%

See 3 more Smart Citations

A Survey on Multi-Label Data Stream Classification

Xiu-lin

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Nowadays, many real-world applications of our daily life generate massive volume of streaming data at a higher speed than ever before, to name a few, Web clicking data streams, sensor network data and credit transaction streams. Contrary to traditional data mining using static datasets, there are several challenges for data stream mining, for instance, finite memory, one-pass and timely reaction. In this survey, we provide a comprehensive review of existing multi-label streams mining algorithms and categorize these methods based on different perspectives, which mainly focus on the multi-label data stream classification. We first briefly summarize existing multi-label and data stream classification algorithms and discuss their merits and demerits. Secondly, we identify mining constraints on classification for multi-label streaming data, and present a comprehensive study in algorithms for multi-label data stream classification. Finally, several challenges and open issues in multi-label data stream classification are discussed, which are worthwhile to be pursued by the researchers in the future. INDEX TERMS Data stream mining, multi-label data, multi-label classification.

show abstract

Section: ) Pt Based Multi-label Data Stream Classification Methodsmentioning

confidence: 99%

Section: ) Ensemble Multi-label Data Stream Classification Methodsmentioning

confidence: 99%

Section: Synthetic Datasetsmentioning

confidence: 99%

Section: Synthetic Datasetsmentioning

confidence: 99%

See 2 more Smart Citations

A Survey on Multi-Label Data Stream Classification

Xiu-lin

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Next, Shi et al (2014a) proposed an efficient and effective method to detect concept drift based on label grouping and entropy for multi-label data, where the labels are grouped by using clustering and association rules. This allowed for an effective detection of concept drift which takes into account label dependence.…”

Section: Multi-label Classification On Data Streamsmentioning

confidence: 99%

Multi-label classification via multi-target regression on data streams

2016

View full text Add to dashboard Cite

Multi-label classification (MLC) tasks are encountered more and more frequently in machine learning applications. While MLC methods exist for the classical batch setting, only a few methods are available for streaming setting. In this paper, we propose a new methodology for MLC via multi-target regression in a streaming setting. Moreover, we develop a streaming multi-target regressor iSOUP-Tree that uses this approach. We experimentally compare two variants of the iSOUP-Tree method (building regression and model trees), as well as ensembles of iSOUP-Trees with state-of-the-art tree and ensemble methods for MLC on data streams. We evaluate these methods on a variety of measures of predictive performance (appropriate for the MLC task). The ensembles of iSOUP-Trees perform significantly better on some of these measures, especially the ones based on label ranking, and are not significantly worse than the competitors on any of the remaining measures. We identify the thresholding problem for the task of MLC on data streams as a key issue that needs to be addressed in order to obtain even better results in terms of predictive performance.

show abstract