Research on Parameters of Affinity Propagation Clustering

  • Bin Gui
  • Xiaoping Yang
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 260)


The affinity propagation clustering is a new clustering algorithm. The volatility is introduced to measure the degree of the numerical oscillations. The research focuses on two main parameters of affinity propagation: preference and damping factor, and considers their relation with the numerical oscillating and volatility, and we find that the volatility can be reduced by increasing the damping factor or preference, which provides the basis for eliminating the numerical oscillating.


Affinity propagation Damping factor Preference Volatility 



This research was supported by the grants from the Natural Science Foundation of China (No. 71271209); Huaiyin Normal University Youth Talents Support Project (NO. 11HSQNZ18).


  1. 1.
    Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Science 315(5814):972–976MathSciNetCrossRefMATHGoogle Scholar
  2. 2.
    Frey BJ, Dueck D (2008) Response to comment on “clustering by passing messages between data points”. Science 319(5864):726CrossRefGoogle Scholar
  3. 3.
    Leone M, Sumedha S, Weigt M (2007) Clustering by soft-constraint affinity propagation: applications to gene-expression data. Bioinformatics 23(20):2708–2715CrossRefGoogle Scholar
  4. 4.
    Sumedha ML, Weigt M (2008) Unsupervised and semi-supervised clustering by message passing: Soft-constrain affinity propagation. Eur Phys J B 66:125–135CrossRefGoogle Scholar
  5. 5.
    Wang K, Zhang J, Li D, Zhang X, Guo T (2007) Adaptive affinity propagation clustering. J Acta Automatica Sinica, 33(12): 1242–1246, (In Chinese)Google Scholar
  6. 6.
    Yu X, Yu J (2008) Semi-supervised clustering based on affinity propagation algorithm. J Software, 19(11):2803–2813, (In Chinese)Google Scholar
  7. 7.
    Zhang X, Wang W, Nørvåg K, Sebag M (2010) K-AP: generating specified K clusters by efficient affinity propagation. ICDM 2010: 1187–1192Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2014

Authors and Affiliations

  1. 1.School of InformationRemin University of ChinaBeijingChina
  2. 2.School of Computer Science and TechnologyHuaiyin Normal UniversityHuaianChina

Personalised recommendations