Decentralized Clustering by Finding Loose and Distributed Density Cores

Chen, Yewang; Tang, Shengyu; Zhou, Lida; Wang, Cheng; Du, Ji-Xiang; Wang, Tian; Pei, Songwen

doi:10.1016/j.ins.2016.08.009

Cited by 48 publications

(26 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To prove the advantages of DDPA-DP, a series of experiments are designed and simulated, and three typical clustering algorithms DBSCAN [12], DP-Clust [21], DPC-KNN [26], FKNN-DPC [31], Dcore [32], and DCNaN [33] are compared with DDPA-DP in these experiments. In this section, experiments are simulated by MATLAB 2015b, and two main performances are analyzed: the accuracies of all clustering algorithms are compared and analyzed in Section 4.1, and the real-time performances of these algorithms are compared and analyzed in Section 4.2.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…In ref [33], a DP-Clust-based algorithm named DCore was proposed by Chen et al DCore uses a concept of density core to find high-density fields and to determine centers and borders, in which the clusters in sparse area can be detected by mean shift thought. In Dcore, data-driven thought is used to adjust the clusters' distribution; however, the value of is r fixed, and the threshold of determine centers is artificially set which restrict the DCore's performance in nonuniform fields.…”

Section: Existing Researches Of Parameter Optimized Densitymentioning

confidence: 99%

See 1 more Smart Citation

A Data‐Driven Parameter Adaptive Clustering Algorithm Based on Density Peak

Wang

2018

Complexity

View full text Add to dashboard Cite

Clustering is an important unsupervised machine learning method which can efficiently partition points without training data set. However, most of the existing clustering algorithms need to set parameters artificially, and the results of clustering are much influenced by these parameters, so optimizing clustering parameters is a key factor of improving clustering performance. In this paper, we propose a parameter adaptive clustering algorithm DDPA-DP which is based on density-peak algorithm. In DDPA-DP, all parameters can be adaptively adjusted based on the data-driven thought, and then the accuracy of clustering is highly improved, and the time complexity is not increased obviously. To prove the performance of DDPA-DP, a series of experiments are designed with some artificial data sets and a real application data set, and the clustering results of DDPA-DP are compared with some typical algorithms by these experiments. Based on these results, the accuracy of DDPA-DP has obvious advantage of all, and its time complexity is close to classical DP-Clust.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Existing Researches Of Parameter Optimized Densitymentioning

confidence: 99%

A Data‐Driven Parameter Adaptive Clustering Algorithm Based on Density Peak

Wang

2018

Complexity

View full text Add to dashboard Cite

show abstract

“…R e ⟵ FindECR(x, k, cid); (9) if R e ≠ ∅ (10) d c (cid) ⟵ max y∈R e d h (y) + b; (11) ECRs cid { } ⟵ R e ; (12) cid ⟵ cid + 1; (13) end if (14) else (15) label(x) ⟵ NOISE; (16) end if (17) (1) R e ⟵ {};…”

Section: Procedures Of the Rnn-dhr Algorithmunclassified

“…Q.enqueue(o); (12) else (13) label(o) ⟵ NOISE; (14) end if (15) end if (16) procedure, those core objects are set to the same cluster number cid, while the others are labeled as noise.…”

Section: Procedures Of the Rnn-dhr Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Local Density Hierarchical Clustering Algorithm Based on Reverse Nearest Neighbors

Liu

Fang

et al. 2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Clustering is widely used in data analysis, and density-based methods are developed rapidly in the recent 10 years. Although the state-of-art density peak clustering algorithms are efficient and can detect arbitrary shape clusters, they are nonsphere type of centroid-based methods essentially. In this paper, a novel local density hierarchical clustering algorithm based on reverse nearest neighbors, RNN-LDH, is proposed. By constructing and using a reverse nearest neighbor graph, the extended core regions are found out as initial clusters. Then, a new local density metric is defined to calculate the density of each object; meanwhile, the density hierarchical relationships among the objects are built according to their densities and neighbor relations. Finally, each unclustered object is classified to one of the initial clusters or noise. Results of experiments on synthetic and real data sets show that RNN-LDH outperforms the current clustering methods based on density peak or reverse nearest neighbors.

show abstract

Task scheduling using edge computing system in smart city

Zheng

Guo

2020

Int J Communication

View full text Add to dashboard Cite

Summary A smart city will become an inevitable trend of social development. It includes many Cyber Physical Systems (CPSs) that can be endowed with different levels of intelligence. With the combination of CPS and cloud computing, smart sensor cloud system becomes popular in many areas where the physical nodes can be shared with multiple users, which can obtain high performance computing. In view of this idea, this paper studies the specific application of smart city‐online task scheduling. We treat each task as a physical node, and the scheduling of multiple computing tasks is to achieve sustainable services. However, when a physical sensor node receives commands from multiple tasks simultaneously, namely, resource conflict management problem. This conflicting resource management problem leads to the failures of services. To solve the resource conflict problem, we propose an extended Hungarian algorithm based on edge computing, which can obtain smaller delay and achieve sustainable service. Edge computing can be viewed as a buffer and controller between CPS and the cloud and can build a sustainable service system by handling malicious attacks. The experimental results show that our proposed method can effectively reduce resource conflicts, improve resource utilization, and make services more sustainable.

show abstract

Decentralized Clustering by Finding Loose and Distributed Density Cores

Cited by 48 publications

References 33 publications

A Data‐Driven Parameter Adaptive Clustering Algorithm Based on Density Peak

A Data‐Driven Parameter Adaptive Clustering Algorithm Based on Density Peak

A Novel Local Density Hierarchical Clustering Algorithm Based on Reverse Nearest Neighbors

Task scheduling using edge computing system in smart city

Contact Info

Product

Resources

About