Genetic algorithm-tuned entropy-based fuzzy C-means algorithm for obtaining distinct and compact clusters

Dey, Vidyut; Pratihar, Dilip Kumar; Datta, G. L.

doi:10.1007/s10700-011-9097-2

Cited by 22 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the utilization of the exponential transformation of other types of distance measures (Xu 2012) in a fuzzy partitioning around medoids procedure or in other typologies of fuzzy clustering, e.g. entropy-based fuzzy clustering (Dey et al 2011) and bi-objective fuzzy clustering (Hung 2007), will be considered for the intervalvalued data case.…”

Section: Final Remarksmentioning

confidence: 99%

Exponential distance-based fuzzy clustering for interval-valued data

D’Urso

Massari

Giovanni

et al. 2016

Fuzzy Optim Decis Making

View full text Add to dashboard Cite

In several real life and research situations data are collected in the form of\ud intervals, the so called interval-valued data. In this paper a fuzzy clustering method\ud to analyse interval-valued data is presented. In particular, we address the problem of\ud interval-valued data corrupted by outliers and noise. In order to cope with the presence\ud of outliers we propose to employ a robust metric based on the exponential distance\ud in the framework of the Fuzzy C-medoids clustering mode, the Fuzzy C-medoids\ud clustering model for interval-valued data with exponential distance. The exponential\ud distance assigns small weights to outliers and larger weights to those points that are\ud more compact in the data set, thus neutralizing the effect of the presence of anomalous\ud interval-valued data. Simulation results pertaining to the behaviour of the proposed\ud approach as well as two empirical applications are provided in order to illustrate the\ud practical usefulness of the proposed method

show abstract

Section: Final Remarksmentioning

confidence: 99%

Exponential distance-based fuzzy clustering for interval-valued data

D’Urso

Massari

Giovanni

et al. 2016

Fuzzy Optim Decis Making

View full text Add to dashboard Cite

show abstract

“…This visualization tool is applicable for 3-D data. For visualization purpose, higherdimensional data are projected or mapped to either 2-D or 3-D [24]. Hence, dimension reduction techniques are applied to this purpose.…”

Section: Visualization Of High Dimensional Clustersmentioning

confidence: 99%

A new semi-supervised clustering technique using multi-objective optimization

2015

View full text Add to dashboard Cite

Semi-supervised clustering techniques have been proposed in the literature to overcome the problems associated with unsupervised and supervised classification. It considers a small amount of labeled data and the whole data distribution during the process of clustering a data. In this paper, a new approach towards semi-supervised clustering is implemented using multiobjective optimization (MOO) framework. Four objective functions are optimized using the search capability of a multiobjective simulated annealing based technique, AMOSA. These objective functions are based on some unsupervised and supervised information. First three objective functions represent, respectively, the goodness of the partitioning in terms of Euclidean distance, total symmetry present in the clusters and the cluster connectedness. For the last objective function, we have considered different external cluster validity indices, including adjusted rand index, rand index, a newly developed min-max distance based MMI index, NMMI index and Minkowski Score. Results show that the proposed semisupervised clustering technique can effectively detect the appropriate number of clusters as well as the appropriate partitioning from the data sets having either well-separated clusters of any shape or symmetrical clusters with or without overlaps. Twenty four artificial and five real-life data sets have been used in the evaluation. We develop five different versions of Semi-GenClustMOO clustering technique by varying the external cluster validity indices. Obtained partitioning results are compared with another recently developed multiobjective semi-supervised clustering technique, Mock-Semi. At the end of the paper the effectiveness of the proposed Semi-GenClustMOO clustering technique is shown in segmenting one remote sensing satellite image on the part from the city of Kolkata.

show abstract

“…To ensure this, a new clustering algorithm may be proposed, in which the cluster centers will be selected initially using the EFC algorithm from a set of data points to be clustered and then, they will be locally (within the cluster) updated to ensure compactness of the clusters using the FCM algorithm. An early version of this new proposal of clustering algorithm has recently been reported by Dey et al (2011), but there is a chance of further improvement of its performance. It will be studied in future.…”

Section: Scope For Future Studymentioning

confidence: 99%

Fuzzy clustering of mechanisms

Ghosh

Pratihar

Amarnath

et al. 2012

Sadhana

Self Cite

View full text Add to dashboard Cite

During the course of development of Mechanical Engineering, a large number of mechanisms (that is, linkages to perform various types of tasks) have been conceived and developed. Quite a few atlases and catalogues were prepared by the designers of machines and mechanical systems. However, often it is felt that a clustering technique for handling the list of large number of mechanisms can be very useful, if it is developed based on a scientific principle. In this paper, it has been shown that the concept of fuzzy sets can be conveniently used for this purpose, if an adequate number of properly chosen attributes (also called characteristics) are identified. Using two clustering techniques, the mechanisms have been classified in the present work and in future, it may be extended to develop an expert system, which can automate type synthesis phase of mechanical design. To the best of the authors' knowledge, this type of clustering of mechanisms has not been attempted before. Thus, this is the first attempt to cluster the mechanisms based on some quantitative measures. It may help the engineers to carry out type synthesis of the mechanisms.

show abstract

Genetic algorithm-tuned entropy-based fuzzy C-means algorithm for obtaining distinct and compact clusters

Cited by 22 publications

References 24 publications

Exponential distance-based fuzzy clustering for interval-valued data

Exponential distance-based fuzzy clustering for interval-valued data

A new semi-supervised clustering technique using multi-objective optimization

Fuzzy clustering of mechanisms

Contact Info

Product

Resources

About