199,603 Hits in 6.9 sec

An extended version of the k-means method for overlapping clustering

Guillaume Cleuziou
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
We show that the problem of finding a suitable coverage of data by overlapping clusters is not a trivial task.  ...  We propose a new objective criterion and the associated algorithm OKM that generalizes the k-means algorithm.  ...  Acknowledgments The research reported in this paper is partially funded by the French National Research Agency ANR (project GD2GS).  ... 
doi:10.1109/icpr.2008.4761079 dblp:conf/icpr/Cleuziou08 fatcat:sylgheldjnbfrkpklayd2bne6y

Phase Transition in a Noise Reduction Model: Shrinking or Percolation? [article]

J. van Mourik, K. Y. Michael Wong, D. Bolle'
1998 arXiv   pre-print
We found that the transition from an extended solution space to a shrunk space is retarded because of the symmetry of the constraints, in contrast to the analogous problem of pattern storage.  ...  A model of noise reduction (NR) for signal processing is introduced. Each noise source puts a symmetric constraint on the space of the signal vector within a tolerable overlap.  ...  When two points in the version space are sampled, they have a high probability to share an overlap q 0 , meaning that they belong to different clusters.  ... 
arXiv:cond-mat/9804115v1 fatcat:nehmi4zf2nezfnma6mpqwlbaxe

Whole Time Series Data Streams Clustering: Dynamic Profiling of the Electricity Consumption

Krzysztof Gajowniczek, Marcin Bator, Tomasz Ząbkowski
2020 Entropy  
Despite the proposal of many clustering techniques applicable for grouping the observations of a single data stream, only a few of them are focused on splitting the whole data streams into the clusters  ...  This work investigates various algorithms (and their improvements) what allows us to formulate the clusters, in real time, based on smart meter data.  ...  For example, the application of the k-means algorithm for clustering of the daily load profiles of individual users was described in [17, [20] [21] [22] .  ... 
doi:10.3390/e22121414 pmid:33333937 pmcid:PMC7765420 fatcat:5oi5x36za5dsnmli7c53zdwq4e

Map-guided Hyperspectral Image Superpixel Segmentation Using Proportion Maps [article]

Hao Sun, Alina Zare
2017 arXiv   pre-print
The proposed approach develops a hyperspectral-appropriate version of the SLIC superpixel segmentation algorithm, leverages map information to guide segmentation, and incorporates the semi-supervised Partial  ...  A map-guided superpixel segmentation method for hyperspectral imagery is developed and introduced.  ...  During the final stage, K = 7 was used during K-means clustering. The superpixel segmentation results for the proposed method and comparison algorithms are shown in Fig. 5 .  ... 
arXiv:1701.01745v1 fatcat:ygfeng25pzdobefytxh27so6q4

"Anti-Bayesian" flat and hierarchical clustering using symmetric quantiloids

Hugo Lewi Hammer, Anis Yazidi, B. John Oommen
2017 Information Sciences  
Within the domain of clustering, the Bayesian principle corresponds to assigning the unlabelled samples to the cluster whose mean (or centroid) is the closest.  ...  This paper, extends the results of [1] in many directions.  ...  In this paper, we have extended the first-reported AB clustering methods proposed in [1] . This paper has extended the results of [1] in many directions.  ... 
doi:10.1016/j.ins.2017.08.017 fatcat:nyenjhhffjax5lwozlvhcigioa

A Novel Fuzzy Clustering Approach for Gene Classification

Meskat Jahan, Mahmudul Hasan
2020 International Journal of Advanced Computer Science and Applications  
MH Extended K-Means cluster algorithm which is a nonparametric extension of the traditional K-Means algorithm and provides the solution for automatic cluster detection including runtime cluster selection  ...  In the end, MH clustering and MH Extended K-Means clustering algorithms were found as a triumph over traditional algorithms.  ...  MH Extended K-Means Clustering Algorithm The MH Extended K-Means clustering algorithm is an extension of the K-Means algorithm.  ... 
doi:10.14569/ijacsa.2020.0110809 fatcat:ha4xlttyazaujkmi6qgj26ncha

Non-exhaustive, Overlappingk-means [chapter]

Joyce Jiyoung Whang, Inderjit S. Dhillon, David F. Gleich
2015 Proceedings of the 2015 SIAM International Conference on Data Mining  
Furthermore, by considering an extension to weighted kernel k-means, we can tackle the case of non-exhaustive and overlapping graph clustering.  ...  our algorithm outperforms state-of-the-art overlapping community detection methods.  ...  There have been many attempts to extend k-means to overlapping clustering. For example, [4] defines OKM.  ... 
doi:10.1137/1.9781611974010.105 dblp:conf/sdm/WhangDG15 fatcat:jzfql4xoerc3hhuqgx4e6lu2pe

Constrained Text Coclustering with Supervised and Unsupervised Constraints

Yangqiu Song, Shimei Pan, Shixia Liu, Furu Wei, Michelle X. Zhou, Weihong Qian
2013 IEEE Transactions on Knowledge and Data Engineering  
The results of our evaluation over two benchmark data sets demonstrate the superiority of our approaches against a number of existing approaches.  ...  We then use an alternating expectation maximization (EM) algorithm to optimize the model.  ...  ACKNOWLEDGMENTS The authors would like to thank Fei Wang for insightful discussions and help on implementing the constrained NMF algorithms.  ... 
doi:10.1109/tkde.2012.45 fatcat:55s3hvhkqjfvxcmc3jsz43czvm

The Iterative Extraction Approach to Clustering [chapter]

Boris Mirkin
2008 Lecture Notes in Computational Science and Engineering  
Specifically, two ITEX derived clustering methods, iK-Means and ADDI-S, are presented as well as update results on theoretical, experimental and applicational aspects of these methods.  ...  The Iterative Extraction approach (ITEX) extends the one-by-one extraction techniques in Principal Component Analysis to other additive data models.  ...  K-Means and iK-Means clustering K-Means is one of the most popular clustering methods.  ... 
doi:10.1007/978-3-540-73750-6_6 fatcat:sqrj6qrt6nbq3oxt3go2baycpu

Fuzzy C-means++: Fuzzy C-means with effective seeding initialization

Adrian Stetco, Xiao-Jun Zeng, John Keane
2015 Expert systems with applications  
Fuzzy C-means has been utilized successfully in a wide range of applications, extending the clustering capability of the K-means to datasets that are uncertain, vague and otherwise hard to cluster.  ...  This paper introduces the Fuzzy C-means++ algorithm which, by utilizing the seeding mechanism of the K-means++ algorithm, improves the effectiveness and speed of Fuzzy C-means.  ...  Acknowledgment Stetco acknowledges the support of an EPSRC doctoral training grant; Keane acknowledges an IBM Faculty Award.  ... 
doi:10.1016/j.eswa.2015.05.014 fatcat:pqjbd572l5gnjih75xa2giib44

Multi-class Spectral Clustering with Overlaps for Speaker Diarization [article]

Desh Raj and Zili Huang and Sanjeev Khudanpur
2020 arXiv   pre-print
Given an overlap detector and a speaker embedding extractor, our method performs spectral clustering of segments informed by the output of the overlap detector.  ...  This paper describes a method for overlap-aware speaker diarization.  ...  ACKNOWLEDGMENT The authors thank Takuya Yoshioka for providing simulation scripts for the LibriCSS training data, and Leibny Paola García-Perera for helpful discussions and insights.  ... 
arXiv:2011.02900v1 fatcat:lid7viypejhk7iqdfg4l5skz4e

On Selecting an Optimal Number of Clusters for Color Image Segmentation

Hoel Le Capitaine, Carl Frelicot
2010 2010 20th International Conference on Pattern Recognition  
This paper addresses the problem of region-based color image segmentation using a fuzzy clustering algorithm, e.g. a spatial version of fuzzy c-means, in order to partition the image into clusters corresponding  ...  Experimental results and comparison with other existing methods show the validity and the efficiency of the proposed method.  ...  An overlap measure between l fuzzy clusters for each point x k of X described by its membership vector u k can be obtained by (3) .  ... 
doi:10.1109/icpr.2010.827 dblp:conf/icpr/CapitaineF10a fatcat:usev7wdpyzer7kchbirw7bqete

Spectral Clustering of Mixed-Type Data

Felix Mbuga, Cristina Tortora
2021 Stats  
The new method includes an automatic tuning of the variable weight and kernel parameter.  ...  The performance of spectral clustering in different scenarios is compared with that of two state-of-the-art mixed-type data clustering methods, k-prototypes and KAMILA, using several simulated and real  ...  Acknowledgments: The authors want to thank the two anonymous reviewers. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/stats5010001 fatcat:bajxjcxpljc45cn3wsshyrb2py

Reducing Program Comprehension Effort in Evolving Software by Recognizing Feature Implementation Convergence

J. Kothari, T. Denton, A. Shokoufandeh, S. Mancoridis
2007 15th IEEE International Conference on Program Comprehension (ICPC '07)  
The implementations of software features evolve as an application matures.  ...  selecting few key features that give an overview of the system.  ...  The overlap is shown for four versions of the software system.  ... 
doi:10.1109/icpc.2007.33 dblp:conf/iwpc/KothariDSM07 fatcat:c3eszy2rmrebpbac5xbnt7zlmu

Overlapping community detection in networks based on link partitioning and partitioning around medoids [article]

Alexander Ponomarenko, Leonidas Pitsoulis, Marat Shamshetdinov
2021 arXiv   pre-print
For small and medium-size networks, the exact solution was found, while for large networks we found solutions with a heuristic version of the LPAM method.  ...  In this paper, we present a new method for detecting overlapping communities in networks with a predefined number of clusters called LPAM (Link Partitioning Around Medoids).  ...  Acknowledgments Authors thank Anna Yaushkina and Nikita Putikhin for the help with implementation of amplified commute distance function and for adding heuristic version of LPAM method.  ... 
arXiv:1907.08731v2 fatcat:kajk2jf55jczfp53udduzsppym
« Previous Showing results 1 — 15 out of 199,603 results