Algorithms for Clustering Data

By Anil K. Jain, R. Dubes. 1988

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Is data clustering in adversarial settings secure? (2013, background): “Single-linkage hierarchical clustering is a bottom-up algorithm that produces a hierarchy of clusterings, as any other hierarchical agglomerative clustering algorithm [18].”
Understanding Membership Inferences on Well-Generalized Learning Models (2018, methodology): “s 8 connecting the corresponding nodes. This allows us to recast our problem as the k-lightest subgraph problem [37], which is NP-hard. We obtain an approximate solution using hierarchical clustering [20]. For this, we cluster the records into k disjoint clusters based on their pairwise cosine dista”
Maximum Margin Multiple Instance Clustering With Applications to Image and Text Clustering (2011, background): “As another branch of machine learning, clustering [ 7 ] is a research topic in both data mining and machine learning, and plays an important role in applications such as text mining, web analysis, marketing, etc.”
An extensive comparative study of cluster validity indices (2013, background): “Therefore, the purpose of clustering is to identify natural structures in a dataset [1] [2] [3] [4] and it is widely used in many fields such as psychology [5], biology [4], pattern recognition [3], image processing [6] and computer security [7].”
A Systematic Review on Educational Data Mining (2017, methodology): “Further details on these algorithms can be found in the work of Jain & Dubes [23].”
All-atom ab initio folding of a diverse set of proteins. (2006, methodology): “Each REMC simulation takes approximately 10 hours on 30 2.4 GHz Xeon CPUs, which results in about 190 CPU days for each protein.”
Functional clustering and identifying substructures of longitudinal data (2007, methodology): “A comprehensive summary of aRand can be found in section 4.4.1 of Jain and Dubes (1988), and in section 2.1 of Yeung and Ruzzo (2001) and their supplementary material.”
GA-Net: A Genetic Algorithm for Community Detection in Social Networks (2008, methodology): “The method is a divisive hierarchical clustering method [6] based on an iterative removal of edges from the network.”
Image classification for content-based indexing (2001, methodology): “This method differs from traditional bootstrapping [ 11 ] which assumes that the original training samples are available for sampling with replacement.”
Performance Evaluation of Some Clustering Algorithms and Validity Indices (2002, methodology): “The K-Means algorithm [1], [2] is an iterative scheme that evolves K crisp, compact, and hyperspheroidal clusters in the data such that a measure J ¼ X n j¼1 X K k¼1 u kj jjx j À z k jj 2 ð1Þ is minimized.”
Regionalization with dynamically constrained agglomerative clustering and partitioning (REDCAP) (2008, background): “The three agglomerative clustering methods differ in their distance definition (i.e. dissimilarity) between clusters (Jain and Dubes 1988; Hastie et al. 2001).”
Algorithms for hierarchical clustering: an overview (2012, methodology): “Five main steps are involved in the GRIDCLUS method: (1) insertion of points into the grid structure, (2) calculation of density indices, (3) sorting the blocks with respect to their density indices, (4) identification of cluster centers, and (5) traversal of neighbor blocks.”
Noise-robust Soft Clustering of Gene Expression Time-course Data (2005, background): “It has been widely used in numerous fields of scientific research.(1) Clustering can be especially useful if prior knowledge is little or non-existent, since it requires minimal prior assumptions.”
Clustering-Based Hyperspectral Band Selection Using Information Measures (2007, methodology): “Several studies point out that this method outperforms other hierarchical clustering methods [18], but, in our case, the process also helps us to form groups with low variance in their level of similarity.”
A Comprehensive Survey of Clustering Algorithms (2015, methodology): “The complete definition for clustering, however, isn’t come to an agreement, and a classic one is described as follows [1]: (1) Instances, in the same cluster, must be similar as much as possible; (2) Instances, in the different clusters, must be different as much as possible; (3) Measurement for si”
Feature normalization and likelihood-based similarity measures for image retrieval (2001, methodology): “Another normalization procedure is to transform the feature component x to a random variable with zero mean and unit variance as ˜ x = x − µ σ (2) where µ and σ are the sample mean and the sample standard deviation of that feature respectively (Jain and Dubes (1988)).”