Earticle

초록

영어

Normally, improving the performance of clustering depends on improvement of the algorithm. On the basis, this paper presents a hybrid strategy optimization algorithm that K-means algorithm effectively combined with PSO algorithm, which not only has played their respective advantages, but also reflected a hybrid performance. First of all, combined with a semi-supervised clustering idea， to optimize the clustering center of particle by K - means in the iteration of algorithm, enhanced the searching capability of the particles. Secondly, improved the traditional K - means enhance the ability of the algorithm to deal with the concave and convex points. Finally, the algorithm is introduced into the particle state determination mechanism, on implementing mutation for unstable particles, so that the algorithm to obtain stable performance. Experimental results show that the hybrid algorithm optimization ability is outstanding, and the convergence and stability can be effectively improved.

목차

Abstract
 1. Introduction
 2. Improved Semi-Supervised Clustering Algorithm
  2.1 K-Means Algorithm
  2.2 Improve Algorithm
 3. The Clustering Algorithm Combined Number of Particle
  3.1 Particle Swarm Optimization
  3.2 Particle Swarm Optimization
  3.3 Improved Particle
  3.4 Clustering Encoding
  3.5 Algorithm Step
 4. Experimental Analysis
 5. Conclusion
 References

키워드

저자

• Li Yi-ran [ College of Applied Technology, University of Science and Technology Liaoning, Anshan Liaoning 114011, China ]
• Zhang Chun-na [ School of Software, University of Science and Technology Liaoning, Anshan Liaoning 114051, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Hybrid Information Technology

• 간기

  격월간

• pISSN

  1738-9968

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605