Fast and Effective Active Clustering Ensemble Based on Density Peak

Shi, Yifan; Yu, Zhiwen; Cao, Wenming; Chen, C. L. Philip; Wong, Hau-San; Han, Guoqiang

doi:10.1109/tnnls.2020.3015795

Cited by 19 publications

(10 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The purpose of EC is to combine multiple base clusterings into a better and more robust consensus clustering [21], [22], [23], [24], [25], [26], [27], [28], [29]. In the final stage of our FastMICE approach, multiple base clusterings are fused into a unified clustering result, which can be viewed as an EC process.…”

Section: Ensemble Clusteringmentioning

confidence: 99%

“…Despite the progress of these EC works [21], [22], [23], [24], [25], [26], [27], [28], [29], most of them are devised for single-view datasets and lack the consideration of multiview scenarios. Recently Tao et al [31] proposed a multiview ensemble clustering (MVEC) method, which learns a consensus clustering from the multiple co-association matrices built in multiple views with low-rank and sparse constraints.…”

Section: Ensemble Clusteringmentioning

confidence: 99%

See 1 more Smart Citation

Fast Multi-view Clustering via Ensembles: Towards Scalability, Superiority, and Simplicity

Dong¹,

Wang²,

Lai³

2022

Preprint

View full text Add to dashboard Cite

Despite significant progress, there remain three limitations to the previous multi-view clustering algorithms. First, they often suffer from high computational complexity, restricting their feasibility for large-scale datasets. Second, they typically fuse multi-view information via one-stage fusion, neglecting the possibilities in multi-stage fusions. Third, dataset-specific hyperparameter-tuning is frequently required, further undermining their practicability. In light of this, we propose a fast multi-view clustering via ensembles (FastMICE) approach. Particularly, the concept of random view groups is presented to capture the versatile view-wise relationships, through which the hybrid early-late fusion strategy is designed to enable efficient multi-stage fusions. With multiple views extended to many view groups, three levels of diversity (w.r.t. features, anchors, and neighbors, respectively) are jointly leveraged for constructing the view-sharing bipartite graphs in the early-stage fusion. Then, a set of diversified base clusterings for different view groups are obtained via fast graph partitioning, which are further formulated into a unified bipartite graph for final clustering in the late-stage fusion. Remarkably, FastMICE has almost linear time and space complexity, and is free of dataset-specific tuning. Experiments on twenty multi-view datasets demonstrate its advantages in scalability (for extremely large datasets), superiority (in clustering performance), and simplicity (to be applied) over the state-of-the-art.

show abstract

Section: Ensemble Clusteringmentioning

confidence: 99%

Section: Ensemble Clusteringmentioning

confidence: 99%

Fast Multi-view Clustering via Ensembles: Towards Scalability, Superiority, and Simplicity

Dong¹,

Wang²,

Lai³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Wang et al proposed an active learning method based on density clustering (ALEC) that takes the structure of the data into account and selects the most representative instances [25]. And then, Shi et al proposed an active density peak (ADP) clustering algorithm that considers both representativeness and informativeness, while informative instances are queried to reduce the uncertainty of clustering results [26]. These algorithms are based on clustering to ensure the minimum distance between two samples.…”

Section: Activementioning

confidence: 99%

AS-DMF: A Lightweight Malware Encrypted Traffic Detection Method Based on Active Learning and Feature Selection

Huo

Zhao

Zhang

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The sharp increasing volume of encrypted traffic generated by malware brings a huge challenge to traditional payload-based malicious traffic detection methods. Solutions that based on machine learning and deep learning are becoming mainstream. However, the machine learning-based methods are limited by manual-design features, which have the problem of highly correlated multicollinearity. And both methods rely heavily on a large number of labeled samples, which needs lots of human effort. In this paper, we apply the active learning to the malicious encrypted traffic detection problem and propose AS-DMF framework. AS-DMF is a lightweight detection framework that combine the uncertainty sampling and density-based query strategy to query the informative and representative instances from the sample set and then train them in a detection (DMF) model. Moreover, we propose a feature selection mechanism which can select the meaningful features of traffic efficiently. Our comprehensive experiments on the real-word dataset indicate that AS-DMF achieves lightweighting at both feature and data levels with a high performance of 0.9460 mAcc.

show abstract

“…In Equation ( 25), the learning rate ρ (m) is sampled from N (0, I) and the symbol * denotes the element-wise multiplication operator. µ z and σ z are learned by the GCN formulated in Equation (14).…”

Section: Learning Algorithmmentioning

confidence: 99%

“…The base clustering members can be produced in many ways, such as by using different clustering algorithms, employing one clustering algorithm with different parameters, and utilizing subsets of data or features [13]. The typical methods used to design a consensus function include a relabeling strategy [14], feature-based methods [15,16], pairwise-similarity-based algorithms [17,18], and graph-based approaches [19,20]. In general, three ensemble-information matrices [11] can be acquired from the base clustering results: the label-assignment matrix, pairwise similarity matrix, and binary cluster association matrix.…”

Section: Introductionmentioning

confidence: 99%

A Clustering Ensemble Framework with Integration of Data Characteristics and Structure Information: A Graph Neural Networks Approach

Wang

2022

Mathematics

View full text Add to dashboard Cite

Clustering ensemble is a research hotspot of data mining that aggregates several base clustering results to generate a single output clustering with improved robustness and stability. However, the validity of the ensemble result is usually affected by unreliability in the generation and integration of base clusterings. In order to address this issue, we develop a clustering ensemble framework viewed from graph neural networks that generates an ensemble result by integrating data characteristics and structure information. In this framework, we extract structure information from base clustering results of the data set by using a coupling affinity measure After that, we combine structure information with data characteristics by using a graph neural network (GNN) to learn their joint embeddings in latent space. Then, we employ a Gaussian mixture model (GMM) to predict the final cluster assignment in the latent space. Finally, we construct the GNN and GMM as a unified optimization model to integrate the objectives of graph embedding and consensus clustering. Our framework can not only elegantly combine information in feature space and structure space, but can also achieve suitable representations for final cluster partitioning. Thus, it can produce an outstanding result. Experimental results on six synthetic benchmark data sets and six real world data sets show that the proposed framework yields a better performance compared to 12 reference algorithms that are developed based on either clustering ensemble architecture or a deep clustering strategy.

show abstract

Fast and Effective Active Clustering Ensemble Based on Density Peak

Cited by 19 publications

References 61 publications

Fast Multi-view Clustering via Ensembles: Towards Scalability, Superiority, and Simplicity

Fast Multi-view Clustering via Ensembles: Towards Scalability, Superiority, and Simplicity

AS-DMF: A Lightweight Malware Encrypted Traffic Detection Method Based on Active Learning and Feature Selection

A Clustering Ensemble Framework with Integration of Data Characteristics and Structure Information: A Graph Neural Networks Approach

Contact Info

Product

Resources

About