scispace - formally typeset
Search or ask a question
Journal Article

Fuzzy Clustering Algorithms - Different Methodologies and Parameters - A Survey

TL;DR: The paper describes about the general working behavior, the methodologies followed on these approaches and the parameters which affects the performance of classical fuzzy clustering algorithms.
Abstract: Fuzzy clustering algorithms are helpful when there exists a dataset with sub groupings of points having indistinct boundaries and overlap between the clusters. This paper gives an overview of different classical fuzzy clustering algorithm. The fuzzy clustering algorithms can be categorized as classical fuzzy clustering and shape based clustering. The paper describes about the general working behavior, the methodologies followed on these approaches and the parameters which affects the performance of classical fuzzy clustering algorithms.
Citations
More filters
01 Jan 2010
TL;DR: In this paper, two polytopes tied to the clustering of a geometric point set into clusters of prescribed sizes are studied and a tight upper bound on the combinatorial diameter of the partition polytope is derived.
Abstract: Motivated by an application in the consolidation of farmland, we study two polytopes tied to the clustering of a geometric point set into clusters of prescribed sizes. First, we characterize the vertices of the 'gravity polytope' as belonging to clusterings that allow a 'full cell decomposition' of the underlying space such that each cluster lies in its own cell. Hereby we obtain an alternative characterization of power diagrams. We show that a vertex of the gravity polytope (and a corresponding full cell decomposition) can be computed by solving a linear program over the 'partition polytope'. This leads to efficient data classification and prediction algorithms. We then study the edge-structure of the two polytopes and derive a tight upper bound on the combinatorial diameter of the partition polytope. Finally, inspired by our polytopal studies, we devise a combinatorial optimization model for our real-world problem.

14 citations


Cites background from "Fuzzy Clustering Algorithms - Diffe..."

  • ...Here, each data vector has a variable degree of membership in each of the clusters [MTA08]....

    [...]

Proceedings ArticleDOI
26 Sep 2014
TL;DR: In this article, Statistical features of the images are extracted from the pixel map of the image and presented to clustering algorithms, Fuzzy C Means and Subtractive clustering algorithm.
Abstract: The web pages are heterogeneous and unstructured. The heterogeneity is due to the hybrid nature of the documents. The unstructureness is due to either multilingual or multimedia content in the web page. The mining should be independent of the language and software. The objective is when any data or content mining is done on a set of data is chosen to form the basis as done with keywords. If the base data is chosen arbitrarily, it is automatic, whereas some 'knowledge' or 'background' is put in the choice it is adaptive. Statistical features of the images are extracted from the pixel map of the image. The extracted features are presented to clustering algorithms, Fuzzy C Means and Subtractive clustering algorithm. The algorithm classifies the given image as a text or image representation. The accuracy of classification is compared and presented.

1 citations


Additional excerpts

  • ...Pixel-based processing to assess overall content is the focus of the study....

    [...]

Proceedings Article
01 Feb 2014
TL;DR: The content of the web is considered as images and statistical features of the images are extracted and presented to the FCM and subtractive clustering, with similarity metric being Euclidean distance.
Abstract: Web pages now-a-days have different forms and types of content. When the web content is considered they are in the form of pictures, video, audio files and text files in different languages. The present study is aimed at this. The content can be multilingual and heterogeneous. The content of the web is considered as images. Statistical features of the images are extracted. The extracted features are presented to the FCM and subtractive clustering, with similarity metric being Euclidean distance. The accuracy is compared.