A New Fuzzy Cover Approach to Clustering

Chiang, Jung-Hsien; Yue, Shihong; Yin, Zong‐Xian

doi:10.1109/tfuzz.2004.825076

Cited by 17 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 1 says that in the sense of time cost and memory the k-fuzzy measure-based CFI-WRLS has advantages over HLMS at the expense of a bit of accuracy. This fully demonstrates that the k-fuzzy measure-based CFI-WRLS is necessary and feasible in real applications, but these results are not better than the existing best results for this purpose [23].…”

Section: Comparison Of Hlms and K-fuzzy Measure-based Cfi-wrlsmentioning

confidence: 76%

Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

Yue

Yin

2005

Information Fusion

View full text Add to dashboard Cite

Section: Comparison Of Hlms and K-fuzzy Measure-based Cfi-wrlsmentioning

confidence: 76%

Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

Yue

Yin

2005

Information Fusion

View full text Add to dashboard Cite

“…For a set of K hypersphere-shaped clusters, if K mean centers suggested by the K-means algorithm accurately coincide with K geometric centers, the K minimal hyperspheres centered on the K mean centers can accurately partition all data objects. However, the K-means algorithm usually cannot attain its optimum at K geometric centers due to the limitations of its objective function, essentially for density-diverse, size-diverse and noisy data-contained clusters [3,5,9,16,17]. We further examine the objective function in the K-means algorithm by the following two noisy data-contained and object distribution-diverse datasets (see Figure 1).…”

Section: Algorithm Analysismentioning

confidence: 99%

“…Consequently, a reasonable and general assumption is to use the average density of K minimal hyperspheres in place of the objective function in the K-means algorithm. In the FCC algorithm [17], we have demonstrated that when the K minimal covers (e.g., hyperspheres, grids, etc.) centralized on K geometric centers enclose K regular clusters the average density of the K covers must be globally maximal.…”

Section: Algorithm Analysismentioning

confidence: 99%

“…This moving process will be completed in finite steps and thus all online grids independently converge to their individual cluster centers in finite steps. On the other hand, an irregular cluster can approximately consist of a number of regular clusters [3,13,17], and any grid will move to one of these regular clusters. Consequently, the convergence of the G-K-MEANS algorithm is guaranteed.…”

Section: Algorithm Analysismentioning

confidence: 99%

“…Some ideas of the proposed grid-clustering method were adopted in this study. In [17] we presented a new fuzzy cover-based clustering (FCC) algorithm. In the FCC algorithm, the concept of the cover is employed to build up the backbones of the final clusters.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An unsupervised grid-based approach for clustering analysis

Yue

Wang²,

Gao

et al. 2010

Sci. China Inf. Sci.

View full text Add to dashboard Cite

In recent years, the growing volume of data in numerous clustering tasks has greatly boosted the existing clustering algorithms in dealing with very large datasets. The K-means has been one of the most popular clustering algorithms because of its simplicity and easiness in application, but its efficiency and effectiveness for large datasets are often unacceptable. In contrast to the K-means algorithm, most existing grid-clustering algorithms have linear time and space complexities and thus can perform well for large datasets. In this paper, we propose a grid-based partitional algorithm to overcome the drawbacks of the K-means clustering algorithm. This new algorithm is based on two major concepts: 1) maximizing the average density of a group of grids instead of minimizing the minimal square error which is applied in the K-means algorithm, and 2) using gridclustering algorithms to thoroughly reformulate the object-driven assigning in the K-means algorithm into a new grid-driven assigning. Consequently, our proposed algorithm obtains an average speed-up about 10-100 times faster and produces better partitions than those by the K-means algorithm. Also, compared with the K-means algorithm, our proposed algorithm has ability to partition any dataset when the number of clusters is unknown. The effectiveness of our proposed algorithm has been demonstrated through successfully clustering datasets with different features in comparison with the other three typical clustering algorithms besides the K-means algorithm.

show abstract

References

2008

Clustering

View full text Add to dashboard Cite

A New Fuzzy Cover Approach to Clustering

Cited by 17 publications

References 26 publications

Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

Parameter estimation for Choquet fuzzy integral based on Takagi–Sugeno fuzzy model

An unsupervised grid-based approach for clustering analysis

References

Contact Info

Product

Resources

About