A parallel k-means clustering initial center selection and dynamic center correction on GPU

被引:0
|
作者
Kakooei, Mohammad [1 ]
Shahhoseini, Hadi Shahriar [1 ]
机构
[1] Iran Univ Sci & Technol, Dept Elect Engn, Tehran, Iran
来源
2014 22nd Iranian Conference on Electrical Engineering (ICEE) | 2014年
关键词
GPGPU; Initial center; Parallel clustering; Dynamic center correction; CUDA;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
K-means clustering algorithm is a partition based clustering algorithm which has been widely used in data mining applications. This algorithm suffers from an issue, named initial centers selection. This problem significantly effects on the quality and running time of clustering. Several literatures discussed on this problem and try to select the best initial centers to prevent final results from getting into local minimum and inaccurate results. Although initial center selection decreases the total running time, it imposes a time overhead that can be solved by parallel design. In addition, previous solutions didn't consider the algorithm behavior after selecting the initial centers, which is considered by dynamic correction in this work. Graphic Processing Unites has several parallel cores which provide a parallel device for developers. This paper proposes a parallel initial centers selection and dynamic center correction on GPU which is fast, accurate and scalable.
引用
收藏
页码:20 / 25
页数:6
相关论文
共 50 条
  • [31] A Fuzzy K-means Clustering Algorithm Using Cluster Center Displacement
    Chang, Chih-Tang
    Lai, Jim Z. C.
    Jeng, Mu-Der
    JOURNAL OF INFORMATION SCIENCE AND ENGINEERING, 2011, 27 (03) : 995 - 1009
  • [32] A fast k-means clustering algorithm using cluster center displacement
    Lai, Jim Z. C.
    Huang, Tsung-Jen
    Liaw, Yi-Ching
    PATTERN RECOGNITION, 2009, 42 (11) : 2551 - 2556
  • [33] GPU-Based Parallel Implementation of k-means Clustering Algorithm for Image Segmentation
    Karbhari, Shruti
    Alawneh, Shadi
    2018 IEEE INTERNATIONAL CONFERENCE ON ELECTRO/INFORMATION TECHNOLOGY (EIT), 2018, : 52 - +
  • [34] Parallel and accurate k-means algorithm on CPU-GPU architectures for spectral clustering
    He, Guanlin
    Vialle, Stephane
    Baboulin, Marc
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2022, 34 (14):
  • [35] Dynamic Incremental K-means Clustering
    Aaron, Bryant
    Tamir, Dan E.
    Rishe, Naphtali D.
    Kandel, Abraham
    2014 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND COMPUTATIONAL INTELLIGENCE (CSCI), VOL 1, 2014, : 308 - 313
  • [36] Fuzzy K-Means Incremental Clustering Based on K-Center and Vector Quantization
    Li, Taoying
    Chen, Yan
    JOURNAL OF COMPUTERS, 2010, 5 (11) : 1670 - 1677
  • [37] A GPU-accelerated parallel K-means algorithm
    Cuomo, S.
    De Angelis, V.
    Farina, G.
    Marcellino, L.
    Toraldo, G.
    COMPUTERS & ELECTRICAL ENGINEERING, 2019, 75 : 262 - 274
  • [38] Data decomposition for parallel K-means clustering
    Gursoy, A
    PARALLEL PROCESSING AND APPLIED MATHEMATICS, 2004, 3019 : 241 - 248
  • [39] Parallel K-Means Clustering Based on MapReduce
    Zhao, Weizhong
    Ma, Huifang
    He, Qing
    CLOUD COMPUTING, PROCEEDINGS, 2009, 5931 : 674 - 679
  • [40] A Median based External Initial Centroid Selection Method for K-means Clustering
    SampathPremkumar, M.
    Ganesh, S. Hari
    2017 2ND WORLD CONGRESS ON COMPUTING AND COMMUNICATION TECHNOLOGIES (WCCCT), 2017, : 143 - 146