K-Means Optimization Algorithm Based on Tightness Mutation

  • Tie Fei LiEmail author
  • Jian Fei Ma
  • Rui Xin Yang
  • Di Wu
  • Yan Guang Shen
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 980)


The random initial clustering center is K - means algorithm to determine the center of traditional way, there will be a clustering result is not stable, optimized to determine the initial clustering center K - means algorithm requires some parameter values, artificial subjective makes the clustering results. Therefore, based on the compactness information of the sample distribution in space, this paper proposes a k-means algorithm to optimize the initial clustering center by using the compactness mutation. The algorithm using the sample space distribution information, through calculating the tightness of the spatial distribution of the sample whether mutations sample information intensity, and based on class sample tightness mutation segmentation principle of cluster center, puts forward the tightness mutation to optimize the initial clustering center of the K - Means algorithm, through the optimization of K - Means clustering algorithm in the UCI machine learning database data set Sentiment labelled sentences and Sentence experiments on Corpus show that the algorithm not only can get better clustering results, The clustering results have high stability.


Clustering K - means algorithm Density mutation Initial cluster center 



This work is supported by Research Projects of Science and Technology in Hebei Higher Education Institutions (ZD2018087, ZD2016017). The authors also gratefully acknowledge the helpful comments and suggestions of the reviewers, which have improved the presentation.


  1. 1.
    Lu, Y.: Research and application of clustering analysis data mining methods. Anhui University (2007)Google Scholar
  2. 2.
    Dongming, T.: Cluster analysis and its application research. University of electronic science and technology (2010)Google Scholar
  3. 3.
    Zhang, J.P., Liu, X.Y.: Research and application of k-means algorithm based on cluster analysis. Comput. Appl. Res. 24(5), 166–168 (2007)Google Scholar
  4. 4.
    Hartigan, J.A.: A k-means clustering algorithm. Appl. Stat. 28(1), 100–108 (1979)CrossRefGoogle Scholar
  5. 5.
    Shen, Y., Xu, H., Liu, H., et al.: Plant depth detection image restoration based on k-means and k-nearest neighbor algorithm. Trans. Chin. Soc. Agric. Eng. 32, 188–194 (2016)Google Scholar
  6. 6.
    Li, M., Zhu, Y., Chen, G., et al.: Kernel k-means clustering algorithm based on local density, a Kernel k-means algorithm based on local density. Comput. Appl. Res. 28(1), 78–80 (2011)Google Scholar
  7. 7.
    Lee, W.H., Lee, S., An, D.U.: Study of a reasonable initial center selection method applied to a k-means clustering. IEICE Trans. Inf. Syst. 96(8), 1727–1733 (2013)CrossRefGoogle Scholar
  8. 8.
    Khan, S., Ahmad, A.: Cluster center initialization algorithm for k-means clustering. Elsevier Science Inc. (2004)Google Scholar
  9. 9.
    Ahmad, A., Hashmi, S.: K-harmonic means type clustering algorithm for mixed datasets. Appl. Soft Comput. 48, 39–49 (2016)CrossRefGoogle Scholar
  10. 10.
    Qian, X., Huang, X.J., Wu, L.D., et al.: A spectral method of K - means initialization initialize K - means of spectrum method. J. Autom. 33(4), 342–346 (2007)zbMATHGoogle Scholar
  11. 11.
    Fu ming Si: Research on an incremental k-means clustering algorithm based on density. J. Changchun Inst. Eng. (Nat. Sci. Edn.) 21(2), 114–117 (2016)Google Scholar
  12. 12.
    Xue, W., Yang, R., Zhao, N., et al.: K-means algorithm for spatial density similarity measurement. Small Microcomput. Syst. 39(1), 53–57 (2018)Google Scholar
  13. 13.
    Xiong, K., Peng, J., Yang, X., et al.: K-means clustering optimization based on kernel density estimation. Comput. Technol. Dev. 27(2), 1–5 (2017)Google Scholar
  14. 14.
    Qi, H.L., Gu, L.: Initial central method of k-means algorithm based on density and minimum distance. Comput. Technol. Dev. 27(9), 60–63, 69 (2017) Google Scholar
  15. 15.
    Ma, F.M., Lu, R.Q., Zhang, T.F.: Rough k-means clustering algorithm based on local density adaptive measurement. Comput. Eng. Sci. 40(1), 184–190 (2018)Google Scholar
  16. 16.
    Lv, H., Huang, X.L., Yang, L.F., et al.: A k-means clustering algorithm based on SIFT feature distribution. In: National Doctoral Academic Conference (2012)Google Scholar
  17. 17.
    Chen, D.N., Cui, X.: K-means initial clustering center selection algorithm based on cell space. Digit. Technol. Appl. (10), 118–119 (2011)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Tie Fei Li
    • 1
    Email author
  • Jian Fei Ma
    • 1
  • Rui Xin Yang
    • 1
  • Di Wu
    • 1
  • Yan Guang Shen
    • 1
  1. 1.School of Information and Electrical EngineeringHebei University of EngineeringHandanChina

Personalised recommendations